Groupement de Recherche en Économie Quantitative d’Aix-Marseille

We analyze the relationship between inequality and economic growth from two directions. The first part of the survey examines the effect of inequality on growth, showing that when capital markets are imperfect, there is not necessarily a trade-off between equity and efficiency. It therefore provides an explanation for two recent empirical findings, namely, the negative impact of inequality and the positive effect of redistribution upon growth. The second part analyzes several mechanisms whereby growth may increase wage inequality, both across and within education cohorts. Technical change, and in particular the implementation of “General Purpose Technologies,” stands as a crucial factor in explaining the recent upsurge in wage inequality.

We study the convergence properties of an alternating proximal minimization algorithm for nonconvex structured functions of the type: L(x,y)=f(x)+Q(x,y)+g(y), where f and g are proper lower semicontinuous functions, defined on Euclidean spaces, and Q is a smooth function that couples the variables x and y. The algorithm can be viewed as a proximal regularization of the usual Gauss-Seidel method to minimize L. We work in a nonconvex setting, just assuming that the function L satisfies the Kurdyka-Łojasiewicz inequality. An entire section illustrates the relevancy of such an assumption by giving examples ranging from semialgebraic geometry to “metrically regular” problems. Our main result can be stated as follows: If L has the Kurdyka-Łojasiewicz property, then each bounded sequence generated by the algorithm converges to a critical point of L. This result is completed by the study of the convergence rate of the algorithm, which depends on the geometrical properties of the function L around its critical points. When specialized to [Formula: see text] and to f, g indicator functions, the algorithm is an alternating projection mehod (a variant of von Neumann's) that converges for a wide class of sets including semialgebraic and tame sets, transverse smooth manifolds or sets with “regular” intersection. To illustrate our results with concrete problems, we provide a convergent proximal reweighted ℓ 1 algorithm for compressive sensing and an application to rank reduction problems.

International audience

Terror as a Bargaining Instrument: A Case Study of Dowry Violence in Rural India by Francis Bloch and Vijayendra Rao. Published in volume 92, issue 4, pages 1029-1043 of American Economic Review, September 2002

This handbook represents the frontier of research into the economics of networks: how and why they form, how they influence behavior, how they help govern outcomes in an interactive world, and how they shape collective decision making, opinion formation, and diffusion dynamics. From a methodological perspective, the authors devote attention to theory, field experiments, laboratory experiments, and econometrics. Theoretical work in network formation, games played on networks, repeated games, and the interaction between linking and behavior are synthesized. A number of chapters are devoted to studying social processes mediated by networks. Topics here include opinion formation, diffusion of information and disease, and learning. There are also chapters devoted to financial contagion and systemic risk. Next, the handbook includes a section that discusses communities more generally, with applications including social trust, favor exchange, and social collateral; the importance of communities for migration patterns, and the role that networks and communities play in the labor market. A prominent role of networks, from an economic perspective, is that they mediate trade. Several chapters cover bilateral trade in networks, strategic intermediation, and the role of networks in international trade. The handbook also discusses the role of networks for organizations. One chapter discusses the role of networks for the performance of organizations, while two other chapters discuss managing networks of consumers and pricing in the presence of network-based spillovers. Finally, the handbook covers the Internet as a network, with attention to the issue of net neutrality.

In practice, bootstrap tests must use a finite number of bootstrap samples. This means that the outcome of the test will depend on the sequence of random numbers used to generate the bootstrap samples, and it necessarily results in some loss of power. We examine the extent of this power loss and propose a simple pretest procedure for choosing the number of bootstrap samples so as to minimize experimental randomness. Simulation experiments suggest that this procedure will work very well in practice.

Udgivelsesdato: June

Using 50,000 tests published in the AER, JPE, and QJE, we identify a residual in the distribution of tests that cannot be explained solely by journals favoring rejection of the null hypothesis. We observe a two-humped camel shape with missing p-values between 0.25 and 0.10 that can be retrieved just after the 0.05 threshold and represent 10–20 percent of marginally rejected tests. Our interpretation is that researchers inflate the value of just-rejected tests by choosing “significant” specifications. We propose a method to measure this residual and describe how it varies by article and author characteristics. (JEL A11, C13)

BACKGROUND: Community-led total sanitation (CLTS) uses participatory approaches to mobilise communities to build their own toilets and stop open defecation. Our aim was to undertake the first randomised trial of CLTS to assess its effect on child health in Koulikoro, Mali. METHODS: We did a cluster-randomised trial to assess a CLTS programme implemented by the Government of Mali. The study population included households in rural villages (clusters) from the Koulikoro district of Mali; every household had to have at least one child aged younger than 10 years. Villages were randomly assigned (1:1) with a computer-generated sequence by a study investigator to receive CLTS or no programme. Health outcomes included diarrhoea (primary outcome), height for age, weight for age, stunting, and underweight. Outcomes were measured 1·5 years after intervention delivery (2 years after enrolment) among children younger than 5 years. Participants were not masked to intervention assignment. The trial is registered with ClinicalTrials.gov, number NCT01900912. FINDINGS: We recruited participants between April 12, and June 23, 2011. We assigned 60 villages (2365 households) to receive the CLTS intervention and 61 villages (2167 households) to the control group. No differences were observed in terms of diarrhoeal prevalence among children in CLTS and control villages (706 [22%] of 3140 CLTS children vs 693 [24%] of 2872 control children; prevalence ratio [PR] 0·93, 95% CI 0·76-1·14). Access to private latrines was almost twice as high in intervention villages (1373 [65%] of 2120 vs 661 [35%] of 1911 households) and reported open defecation was reduced in female (198 [9%] of 2086 vs 608 [33%] of 1869 households) and in male (195 [10%] of 2004 vs 602 [33%] of 1813 households) adults. Children in CLTS villages were taller (0·18 increase in height-for-age Z score, 95% CI 0·03-0·32; 2415 children) and less likely to be stunted (35% vs 41%, PR 0·86, 95% CI 0·74-1·0) than children in control villages. 22% of children were underweight in CLTS compared with 26% in control villages (PR 0·88, 95% CI 0·71-1·08), and the difference in mean weight-for-age Z score was 0·09 (95% CI -0·04 to 0·22) between groups. In CLTS villages, younger children at enrolment (<2 years) showed greater improvements in height and weight than older children. INTERPRETATION: In villages that received a behavioural sanitation intervention with no monetary subsidies, diarrhoeal prevalence remained similar to control villages. However, access to toilets substantially increased and child growth improved, particularly in children <2 years. CLTS might have prevented growth faltering through pathways other than reducing diarrhoea. FUNDING: Bill & Melinda Gates Foundation.

Abstract This chapter addresses the issues of endogenous quantity and endogenous quality of labor, provides a simple model of productivity and wages in cities, and discusses the two main estimation issues. It presents the wage data and the worker-fixed approach to the endogenous quality of labor bias. Cities attract skilled workers so that the effects of skills and urban agglomeration are confounded. Urban agglomeration is a consequence of high local productivity rather than a cause. An instrumental variable approach introduces a new set of geological instruments in addition to standard historical instruments. Furthermore, the chapter shows the results for wages and productivity. Long lags of endogenous explanatory variables make for strong instruments, and geological characteristics are more complicated instruments to play with. The simultaneity problem between employment density and local wages/productivity is relatively small. Better workers are located in more productive areas. This sorting of workers by skills (observed and unobserved) is quantitatively more important than the endogenous quantity of labor bias. The future work would develop more sophisticated approaches to deal with the sorting of workers across places.

BACKGROUND: Dementia is a prevalent condition, affecting 5-7% of people aged 60 years and older, and a leading cause of disability in people aged 60 years and older globally. We aimed to examine the association between alcohol use disorders and dementia risk, with an emphasis on early-onset dementia (<65 years). METHODS: We analysed a nationwide retrospective cohort of all adult (≥20 years) patients admitted to hospital in metropolitan France between 2008 and 2013. The primary exposure was alcohol use disorders and the main outcome was dementia, both defined by International Classification of Diseases, tenth revision discharge diagnosis codes. Characteristics of early-onset dementia were studied among prevalent cases in 2008-13. Associations of alcohol use disorders and other risk factors with dementia onset were analysed in multivariate Cox models among patients admitted to hospital in 2011-13 with no record of dementia in 2008-10. FINDINGS: Of 31 624 156 adults discharged from French hospitals between 2008 and 2013, 1 109 343 were diagnosed with dementia and were included in the analyses. Of the 57 353 (5·2%) cases of early-onset dementia, most were either alcohol-related by definition (22 338 [38·9%]) or had an additional diagnosis of alcohol use disorders (10 115 [17·6%]). Alcohol use disorders were the strongest modifiable risk factor for dementia onset, with an adjusted hazard ratio of 3·34 (95% CI 3·28-3·41) for women and 3·36 (3·31-3·41) for men. Alcohol use disorders remained associated with dementia onset for both sexes (adjusted hazard ratios >1·7) in sensitivity analyses on dementia case definition (including Alzheimer's disease) or older study populations. Also, alcohol use disorders were significantly associated with all other risk factors for dementia onset (all p<0·0001). INTERPRETATION: Alcohol use disorders were a major risk factor for onset of all types of dementia, and especially early-onset dementia. Thus, screening for heavy drinking should be part of regular medical care, with intervention or treatment being offered when necessary. Additionally, other alcohol policies should be considered to reduce heavy drinking in the general population. FUNDING: None.

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et a ̀ la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. 1

In this paper, we model the formation of innovation networks as they emerge from bilateral decisions. In contrast to much of the literature, here firms only consider knowledge production, and not network issues, when deciding on partners. Thus, we focus attention on the effects of the knowledge and information regime on network formation. The effectiveness of a bilateral collaboration is determined by cognitive, relational, and structural embeddedness. Innovation results from the recombination of knowledge held by the partners to the collaboration, and its success is determined in part by the extent to which firms’ knowledge complement each other. Previous collaborations (relational embeddedness) increase the probability of a successful collaboration, as does information gained from common third parties (structural embeddedness). Repeated alliance formation creates a network. Two features are central to the innovation process: how firms pool their knowledge resources, and how firms derive information about potential partners. When innovation is decomposable into separate subtasks, networks tend to be dense; when structural embeddedness is important, networks become cliquish. For some regions in this parameter space, small worlds emerge.

This paper evaluates, in the context of economic geography estimates, the magnitude of the distortions arising from the choice of a specific zoning system, which is also known as the Modifiable Areal Unit Problem (MAUP). We undertake three standard economic geography exercises (the analysis of spatial concentration, agglomeration economies, and trade determinants), using various French zoning systems differentiated according to the size and shape of their spatial units. While size might matter, especially when the dependent variable of a regression is not aggregated in the same way as the explanatory variables and/or the zoning system involves large spatial units, shape does so much less. In any case, both dimensions are of secondary importance compared to specification issues.

Beyond DSGE Models: Toward an Empirically Based Macroeconomics by David Colander, Peter Howitt, Alan Kirman, Axel Leijonhufvud and Perry Mehrling. Published in volume 98, issue 2, pages 236-40 of American Economic Review, May 2008

This article examines, in the light of recent events, the origins of the difficulties that current macroeconomic models have in encompassing the sort of sudden crisis which we are currently observing. The reasons for this are partly due to fundamental problems with the underlying General Equilibrium theory and partly to the unrealistic assumptions on which most financial models are based. What is common to the two is that systematic warnings over more than a century in the case of finance and over 30 years in the case of equilibrium theory have been ignored and we have persisted with models which are both unsound theoretically and incompatible with the data. It is suggested that we drop the unrealistic individual basis for aggregate behaviour and the even more unreasonable assumption that the aggregate behaves like such a ‘rational’ individual. We should rather analyse the economy as a complex adaptive system, and take the network structure that governs interaction into account. Models that do this, of which two examples are given, unlike standard macroeconomic models, may at least enable us to envisage major ‘phase transitions’ in the economy even if we are unlikely to be able to forecast the timing of their onset. (JEL codes: B22, D84, D85, E10, E44)

We summarize a number of regularities that arise in the empirical literature on contractual relationships between manufacturers and their exclusive resellers. We do this using studies of traditional and business‐format‐franchise relationships, as well as studies of sales‐force‐integration decisions. Some of the patterns that we uncover are consistent with a standard incentive‐cum‐insurance theory of organization, while others are not. We briefly review some theoretical extensions that seem promising in terms of reconciling seeming conflicts between theory and practice.

This paper analyzes the relationship between income inequality and inequality of opportunities for income acquisition in nine developed countries during the 1990s. Equality of opportunity is defined as the situation where income distributions conditional on social origin cannot be ranked according to stochastic dominance criteria. We measure social origin by parental education and occupation and use the database built by Roemer et al. (2003 ). Stochastic dominance is assessed using nonparametric statistical tests. Our results indicate strong disparities in the degree of equality of opportunity across countries and a strong correlation between inequality of outcomes and inequality of opportunity. The U.S. and Italy show up as the most unequal countries in terms of both outcome and opportunity. At the opposite extreme, income distributions conditional on social origin are almost the same in Scandinavian countries even before any redistributive policy. We complement the ordinal comparison by resorting to an original scalar “Gini” index of opportunities, which can be decomposed into a risk and a return component. In our sample, inequality of opportunity is mostly driven by differences in mean income conditional on social origin, and differences in risk compensate the return element in most countries.

Simple techniques for the graphical display of simulation evidence concerning the size and power of hypothesis tests are developed and illustrated. Three types of figures—called P value plots, P value discrepancy plots and size–power curves—are discussed. Some Monte Carlo experiments on the properties of alternative forms of the information matrix test for linear regression models and probit models are used to illustrate these figures. Tests based on the outer‐product‐of‐the‐gradient (OPG) regression generally perform much worse in terms of both size and power than efficient score tests.

The Capability Approach (CA) as developed by Amartya Sen and Martha Nussbaum, has in part been a response to the problem of adaptive preferences. Their argument says that people might adapt to certain unfavorable circumstances and any self-evaluation in terms of satisfaction or happiness will in this case necessarily be distorted. To evaluate people's well-being in terms of functionings and capabilities guarantees a more objective picture of people's life. Next to this strong criticism on subjective measurements of well-being, we observe an increasing interest in Subjective Well-Being (SWB) or Happiness studies that are included in the broader field of Hedonic Psychology. In this paper, we thus revise the original critique of adaptive preferences and compare it with a more detailed analysis of adaptation as it is presented in hedonic psychology. It becomes clear that adaptation can be a positive as well as a negative phenomenon and that the adaptive preference critique had a particular narrow view on adaptation. However, this does not mean SWB-research is not any longer susceptible to this critique. An alternative way to assess people's subjective well-being, but which could be considered to be more in line with the CA, is proposed by Daniel Kahneman's Objective Happiness. These are all relatively new considerations, especially in economics. Therefore much more research needs to be done on the positive and negative aspects of adaptation to understand its consequences on well-being – especially when evaluated within the capability-space.