Bootstrapping for standard errors of instrumental variable estimates

Theil and Finke (1983) used the distance from the equator as an instrumental variable (IV) in the estimation of a cross-country demand system. Here we obtain standard errors of the IV estimates using Efron's (1979) bootstrap technique.     

La dimension internationale des politiques urbaines dans le Tiers Monde

Urban policics in dweloping countries are subject to both internal and external determinants. As elsrwhere, internal determinants, social relations and conflicts evolve over urbun space. external determinants are not only related to the general mechanisms of dependency, but also to specific modes of intervention by European and North American countries: such as consultancy firms producing urban studies and plans, lateral cooperative organisms, international institutions, suppliers of key materials for urbanization (cars. buses, pipes), training centres for planners and other technicians, etc. All urban policies in the Third World have this international dimension. It has been said that urban sociology has become the sociology of urban policies. In this case, urban sociology of the Third World has become that of the sociology of urban po1icies in their international dimensions, and it can only be done by examining each end of the chain, each side of the system of relations. Neither side can be studied without the. other in understanding and analysing the phenomenon which links them. The urban sociology of the Third World could provide the opportunity to reconsider from new points or departure, the relations between European or North American roeiologists on the anc hand, and Latin American or African sociologists on the other.     

Virtual research universes and research management

The analysis of large research production, such as universities, towns, cities or even countries, can be performed in several ways. These various approaches can range from the experts' feeling to a more precise analysis of co-authors' productivity. The various results obtained introduce the concept of 'virtual research universes' where the global perception of the production hides the sharp reality of objectives and subjects. The ability to apply the various methodologies and tools to make such analysis can reduce misunderstanding and misleading positions, especially in funding, planning and project evaluation. This paper takes as an example the production of physics laboratories in Marseille (from the INSPEC databae), and shows how the various levels of analysis can move from virtual to real research universes, thus helping the experts in their decision process.     

Another look at the maximum entropy correlation coefficient

A simulation experiment based on 10,000 trials shows that the Fisher transform of the maximum entropy correlation coefficient has a smaller bias than that of the sample correlation coefficient when the parent correlation ρ satisfies |ρ|?0.8, and a smaller root-mean-square error when |ρ|?0.95.     

BIASED TECHNOLOGICAL SHOCKS, RELATIVE WAGE RIGIDITIES AND LOW-SKILLED UNEMPLOYMENT

The contrast between the evolution over the last decades of the European Union (EU) and the US unemployment rates, especially for the low-skilled, is well known. A consensus view is that these different outcomes can be explained by the interactions between common shocks and specific institutional setups. In this paper, we emphasize the interactions between technological changes and wages rigidities. We construct a fully calibrated general equilibrium model with two types of jobs and two types of workers, and with search unemployment. Our simulations show that with wage rigidities, technological changes suffice to generate a continuous rise in the low-skilled unemployment rate and an almost unchanged high-skilled unemployment rate. Without wage rigidities, the unemployment rates remain unchanged but the wage dispersion widens.     

Taxation of multiple greenhouse gases and the effects on income distribution: A case study of the Netherlands

Current economic instruments aimed at climate change mitigation focus on CO₂ emissions only, but the Kyoto Protocol refers to other greenhouse gases (GHG) as well as CO₂. These are CH₄, N₂O, HFCs, PFCs and SF₆. Taxation of multiple greenhouse gases improves the cost-effectiveness of climate change mitigation. It is not yet clear, however, what the effect is of multigas taxation on the distribution of the tax burden across income groups. This paper examines and compares distributional effects of a CO₂ tax and a comprehensive tax that covers all six GHG of the Kyoto Protocol. The study concentrates on the Netherlands in the year 2000. We established tax rates on the basis of marginal abatement cost curves and the Dutch policy target. The distributional effects have been determined by means of environmentally extended input−output analysis and data on consumer expenditures. Our results show that taxation of multiple GHG improves not only the cost-effectiveness of climate change mitigation, but also distributes the tax burden more equally across income groups as compared to a CO₂ tax. These findings are relevant for the debate on the role of non-CO₂ GHG in climate change mitigation.     

New results on rational random behavior

This letter deals with conditions under which it is preferable to buy a zero amount of information; with a proportionality of the information and loss variances associated with two different decision distributions; and with a limiting chi-square distribution of the decision maker's loss when the marginal cost of information is small.     

Maximum likelihood estimation of the Rotterdam model under two different conditions

This letter describes maximum likelihood (ML) procedures for the Rotterdam model (1) when the model is formulated in relative prices and (2) under the constraints of the theory of rational random behavior.     

The positive correlation of affluence and freedom

Non-communist countries rated ‘free’ have on average a much higher per capita income than those rated ‘partly free’ or ‘not free’. The latter two groups differ primarily in terms of the dispersion of their per capita incomes, those rated ‘not free’ being much more dispersed than those rated ‘partly free’.     

The Antonelli matrix and the reciprocal Slutsky matrix

Inverted demand models which describe deflated price changes in terms of quantity changes can be formulated by means of the Antonelli matrix or the reciprocal Slutsky matrix. The two approaches differ with respect to the price deflator and the role of the Divisia volume index.