A Quantile Regression Analysis of Wages in Panama

Differences in the effects of worker characteristics on wages in Panama at different points of the conditional wage distribution are investigated. Public sector employment increases wages relatively more at lower quantiles. Within the public sector, employment in that sector increases wages of the median worker and reduces wage inequality. Presence of a labor union increases relatively more private sector wages at lower quantiles. Unions reduce wage inequality within the union private sector and increase average wages within that sector. In the public sector, the presence of a labor union increases wages of men at lower quantiles at a lower rate than in the private sector. Self-employment decreases wages at lower quantiles and increases wages at higher quantiles. Urban location affects wages in a U-shaped pattern as one moves from lower to higher quantiles. Rates of return to experience are higher for men at higher quantiles. Experience increases men's wage inequality.     

The impact of piped water provision on infant mortality in Brazil: A quantile panel data approach

We examine the impact of piped water on the under-1 infant mortality rate (IMR) in Brazil using a recently developed econometric procedure for the estimation of quantile treatment effects with panel data. The provision of piped water in Brazil is highly correlated with other observable and unobservable determinants of IMR — the latter leading to an important source of bias. Instruments for piped water provision are not readily available, and fixed effects to control for time-invariant correlated unobservables are invalid in the simple quantile regression framework. Using the quantile panel data procedure in Chen and Khan [Chen, S., Khan, S., Semiparametric estimation of non-stationary censored panel model data models with time-varying factor. Econometric Theory 2007; forthcoming], our estimates indicate that the provision of piped water reduces infant mortality by significantly more at the higher conditional quantiles of the IMR distribution than at the lower conditional quantiles (except for cases of extreme underdevelopment). These results imply that targeting piped water intervention toward areas in the upper quantiles of the conditional IMR distribution, when accompanied by other basic public health inputs, can achieve significantly greater reductions in infant mortality.     

State of the “life span revolution” between 1980 and 2000

Using data for 163 countries, state of the “life span revolution” over the period 1980–2000 is studied in terms of measures of cross-country inequality and through least-squares and quantile-regression estimation of simple convergence models. Four main points are noted. First, dynamics of the cross-country distribution of life expectancy during these 20 years seem markedly different from those for the preceding decades: instead of the sharp “convergence” noted until the 1980s, there is lack of convergence and an indication of “divergence”. Second, the divergence is particularly marked during the 1990s. Third, spread of HIV/AIDS has probably been a significant factor in generating divergence during the 1990s. Fourth, besides the sizable temporal heterogeneity, quantile-regression estimates of convergence models reveal a substantial heterogeneity across the top and the bottom quartiles within each period.     

Unemployment duration,Fiscal and monetary policies,and the output gap: How do the quantile relationships look like?

We use country-level panel data on average unemployment duration (AvgUD) to explore how the conditional distribution of AvgUD varies with primary fiscal balance (fiscal policy), short-term interest rate (monetary policy), and the state of the business cycle (output gap). We obtain the quantile estimates via a Bayesian quantile mixed effects model, which has so far received limited attention in economics. Our results reveal rich heterogeneity in both the sign and strength of these associations at different quantiles of AvgUD, in different gender and age groups, and in some transition economies that have markedly different labour markets. A higher fiscal balance/surplus is associated with higher middle-upper quantiles of AvgUD among prime-age and old-age men, whereas the empirical association is weak at all quantiles of AvgUD among young men and women. The association between quantiles of AvgUD and fiscal and monetary policies tends to be weaker among women than men. A better state of the business cycle is associated with a lower AvgUD at all quantiles among young men and women.     

Quantile regression with sample selection: Estimating women's return to education in the U.S.

This study uses quantile regression techniques to analyze changes in the returns to education for women. The data used is the March Current Population Survey for the years 1968, 1973, 1979, 1986 and 1990. The first step in estimating the single (linear) index selection equation uses Ichimura's (1993) semiparametric procedure. To correct for an unknown form of a sample selection bias in the quantile regression, the second step incorporates a nonparametric method, using an idea similar to one developed by Heckman (1980) and Newey (1991) for mean regression, and Buchinsky (1998) for quantile regression.  The results show that: (a) the returns to education increased enormously for the younger cohorts, but very little for the older cohorts; (b) in general the returns are higher at the lower quantiles in the beginning of the sample period and higher at the higher quantiles by the end of the sample period; (c) there is a significant sample selection bias for all age groups at almost all quantiles; (d) toward the end of the sample period there is a significant convergence of the returns at the various quantiles, especially for the younger cohorts and age groups; and (e) the semiparametric estimates of the selection equation are considerably different from those obtained for a parametric probit model.     

Earning functions in Portugal 1982–1994: Evidence from quantile regressions

This paper uses quantile regressions to describe the conditional wage distribution in Portugal and its evolution over the 1980s as well as the implications for increased wage inequality. We find that, although returns to schooling are positive at all quantiles, education is relatively more valued for highly paid jobs. Consequently, schooling has a positive impact on wage inequality. Moreover, this tendency has sharpened over the period. We also find that most of the estimated change in wage inequality was due to changes in the distribution of the worker's attributes, rather than to increased inequality within a particular type of worker. this version: January 2000     

Economic uncertainty and corruption: evidence from a large cross-country data set

A vast amount of research has considered numerous causes and correlates of corruption. Also, there have been many studies of the consequences of various forms of uncertainty. However, exploration of the nexus between economic uncertainty and corruption appears scarce. After providing an intuitive and heuristic linkage between general economic uncertainty and corruption, this article uses a large cross-country data set to augment a fairly standard model with simple proxies for uncertainty and to investigate how economic uncertainty might affect the prevalence of corruption. In addition, a quantile-regression framework is used to judge how the strength of various covariates may differ with the level of corruption. Seven main points emerge from the estimates. First, economic uncertainty is associated positively with corruption, and the relation seems to be robust across measures of uncertainty and corruption. Second, quantile-regression estimates indicate considerable parametric heterogeneity across the distribution of corruption. Third, Gross Domestic Product (GDP) per capita has the expected corruption-mitigating role. Fourth, increased political rights and civil liberties also appear to lower corruption. Fifth, greater government consumption is associated with lower corruption. Sixth, while the hyperinflation dummy lacks significance in most OLS regressions, its significance varies across the distribution of corruption. Seventh, neither police force nor government subsidies shows significance, but transition economies have more corruption.     

Decomposition and wage inequality

Earning differentials are investigated by a quantile regressions based decomposition, which disentangles the inequalities linked to the covariates and coefficients at various quantiles. Gender and region are considered the main sources of inequality. The unexplained gender and regional differences decrease at the highest wages. Their combination at the lower wages’ level affects women more, causing a so-called sticky floor. Gender and regional covariate effects show a prevalence of women covariates compared with the men’s group, and a prevalence of southern women covariates within the women’s group, particularly at the higher quantiles. This can be interpreted as a glass ceiling hindering southern women at higher wages.     

The returns to education and wage penalty from overeducation: New evidence from Vietnam

This study intends to estimate the rate of returns to education in Vietnam, the distributive effects of education on wages, and the wage penalty from the incidence of overeducation in the Vietnamese labor market during 2004–2016. This study employs a pseudo-panel approach to address omitted variables bias and the unconditional quantile regression to identify the heterogeneity of returns to education across the income distribution. Our main finding indicates that the estimated rate of returns to education in Vietnam is approximately 6.5%, showing a downward bias from previous estimates. The returns vary across wage distributions, where a lower rate of return is observed in lower quantiles and a higher rate among those individuals at the higher quantiles. The returns to education have declined since 2008, confirming the oversupply of highly educated workers in the Vietnamese labor market with an estimated wage penalty of 17%. Government assistance measures are needed to reduce the overeducation and the wage penalty issues in the Vietnamese labor market.     

Public Preferences for Hydrogen Buses: Comparing Interval Data, OLS and Quantile Regression Approaches

We use a quantile regression (QR) approach to analyse contingent valuation estimates of public willingness to pay (WTP) for the air and noise pollution reductions associated with the introduction of hydrogen buses in London. QR results show that variables that were not significant in interval regression or ordinary least squares regression become significant at certain quantiles along the WTP distribution. In addition, the determinants of WTP at the lower tail of the distribution differ from those at the higher end of the distribution. Our findings illustrate the usefulness of quantile regression methods for analysing contingent valuation data, enhancing our understanding of the determinants of willingness to pay.     

The Impact of Inheritance on the Distribution of Wealth: Evidence from Great Britain

Using the British Household Panel Survey, we investigate the role of inheritance in shaping the distribution of household wealth in Great Britain during 1995–2005a period characterized by a substantial increase in wealth and an equally important decrease in wealth inequality. Abstracting from behavioral effects, we find that inheritances received during this period accounted for 30 percent of the increase in wealth of inheritors. Regression estimates of the effect of inheritance on wealth accumulation suggest that households spend 30 percent of their inheritances on average, and that there is substantial heterogeneity in household responses. Households that accumulated more wealth saved a larger share of their inheritances, as did middle aged households and those with lower initial wealth. Although inheritances are highly unequal they had a small impact on overall wealth inequality. This mainly reflected the fact that their size relative to other sources of wealth was very small.     

Is aid for trade effective? A panel quantile regression approach

This paper investigates whether Aid for Trade (AfT) leads to greater exports in recipient countries. Using panel quantile regression and two strategies to address endogeneity (AfT lagged by two periods and dynamic OLS), our results suggest that total AfT disbursements promote the export of goods and services mainly for the lower quantiles (0.1, 0.25, 0.50) of the conditional distribution of exports. Hence, countries that export less in volume are those benefitting most from AfT. This effect is mainly driven by the impact on exports of goods rather than on that of services. We also investigate which types of AfT are effective when endogeneity is controlled for. The main results show that whereas aid to improve trade policy and regulation is associated with higher exports for all quantiles, aid used to build infrastructure is found to affect exports at only the lowest tails of the distribution (0.10–0.35) and aid to build productive capacity is generally more effective for the lower quantiles of the export distribution (0.10–0.50). In contrast, aid disbursed for general budget support (an untargeted type of aid) is not associated with greater export levels. This finding holds irrespective of the quantile.     

The return to education and its distribution in urban China: evidence from quantile regression analysis

Based on Chinese Households Income Projects in 2002, this paper discusses the relationship between the return to education and the quantiles of income distribution. The findings in this paper show that the return to education is lower for the higher quantiles, while the estimators also depend on the choice of control variables. The methodology of the quantile regression might be helpful in adjusting the ability bias in the estimation on return to education. The policy implications of the paper highlight the impact of the education expansion in boosting the income growth for those in lower quantiles.     

Schooling and household welfare: The case of Sri Lanka from 1990 to 2006

This paper looks at the effect schooling has had on household welfare in Sri Lanka during the 1990–2006 period, on average and across the welfare distribution. We account for the endogeneity of schooling using quantile instrumental variable estimation as developed in Chernozhukov, Fernández‐Val, and Kowalski ( 2015 ). We use pooled data from four cross‐sectional Household Income Expenditure Surveys. The results show that an extra year of schooling on the part of the most educated adult member in the household can increase welfare (proxied by real per capita consumption expenditure) by 3.8 percent on average. However, the effect varies considerably across the welfare distribution: At the lower end, around the 20th and 25th quantiles, an extra year of education increases welfare by 6 and 5 percent, respectively, while at the median it is around 3.5 percent. At the higher, 90th quantile it is much less, at 1 percent. Thus the marginal effect of schooling on welfare is significant and positive at all levels of the welfare distribution, but highest at the lower and middle quartiles. This result is different from findings in the literature that tend to show larger effects at higher quantiles, when endogeneity is uncorrected.     

US federal redistributive income policies

The purpose of this paper is to present a statiscial model of inequality that allows us to analyse simultaneously the impact of several major redistributive expenditure policies, as well as analysing the impact of income taxation on inequality. The paper includes multivariate exogeeity tests to determine whether the policies have ‘caused’ the observed income distribution or if there is feedback, i.e. inequality in the distribution has lead to enhanced (or diminished) redistributive policies. The direction of causality for these policies has not, to ourr knowledge, been previously examined. This paper examines the usual redistributive vehicles of transfer payments, Social Security, Medicare and Medicaid, as well as analysing the impact of the personal income tax system on inequality. Using time series analysis we can allow for lagged effects of these policies as well as current period effects. This lagged effects specification may be sugggestive of behavioural responses to policy changes.     

Individual heterogeneity in the returns to schooling: instrumental variables quantile regression using twins data

Considerable effort has been exercised in estimating mean returns to education while carefully considering biases arising from unmeasured ability and measurement error. Recent work has investigated whether there are variations from the “mean” return to education across the population with mixed results. We use an instrumental variables estimator for quantile regression on a sample of twins to estimate an entire family of returns to education at different quantiles of the conditional distribution of wages while addressing simultaneity and measurement error biases. We test whether there is individual heterogeneity in returns to education and find that: more able individuals obtain more schooling perhaps due to lower marginal costs and/or higher marginal benefits of schooling and that higher ability individuals (those further to the right in the conditional distribution of wages) have higher returns to schooling consistent with a non-trivial interaction between schooling and unobserved abilities in the generation of earnings. The estimated returns are never lower than 9 percent and can be as high as 13 percent at the top of the conditional distribution of wages but they vary significantly only along the lower to middle quantiles. Our findings may have meaningful implications for the design of educational policies.     

THE DISTRIBUTION OF WEALTH IN CANADA

For estimates of the wealth distribution Canada depends on household surveys taken at 6–7 year intervals. The latest data from this source refer to household balance sheets in the spring of 1977. A comparison with 1970 shows that there is little change in the composition of wealth held by households but that inequality of the wealth distribution has been somewhat reduced. Wealth data by age of family head is presented in order to describe more fully the wealth distribution and composition in Canada.
Weaknesses in the data are discussed as well as the difficulties of making appropriate adjustments to the data at the micro record level. For policy evaluation and formulation purposes the lack of comprehensive estimates inclusive of pension wealth as well as the small sample size (12,700 usable records) have been perceived as greater obstacles to utilizing the data than the underestimate in aggregate assets and debts which affects more the higher than the middle and lower ranges of the wealth distribution.     

The effects of works councils on overtime hours

This study analyzes the relation between works councils and overtime hours in Germany. The estimated effects differ considerably in dependence of standard contracted working time. Furthermore, we find differences across the quantiles of the overtime hours distribution and these differences between quantiles also vary between employees of establishment with and without works councils. By considering contracted working time effects and comparing conditional quantiles we find in some quantiles that employees of establishments with a works council have a higher amount of overtime hours if an employee regularly works 35 hours per week. This effect diminishes with increasing quantiles of the distribution of overtime hours. If an employee of a codetermined establishment regularly works 40 hours per week, overtime hours are reduced. This reduction becomes larger if higher quantiles are analyzed.     

Time to learn? The organizational structure of schools and student achievement

Utilizing parametric and nonparametric techniques, we asses the impact of a heretofore relatively unexplored ‘input’ in the educational process, time allocation, on the distribution of academic achievement. Our results indicate that school year length and the number and average duration of classes affect student achievement. However, the effects are not homogeneous – in terms of both direction and magnitude – across the distribution. We find that test scores in the upper tail of the distribution benefit from a shorter school year, while a longer school year increases test scores in the lower tail. Furthermore, test scores in the lower quantiles increase when students have at least eight classes lasting 46–50 min on average, while test scores in the upper quantiles increase when students have seven classes lasting 45 min or less or 51 min or more.