The fact that a large portion of households does not participate in the stock market is empirically well established (King and Leape 1984, Mankiw and Zeldes 1991, Bertaut and Starr-McCluer 2002). This stands at odds with the classical theory of portfolio choice which predicts that agents should always have a non-zero position in risky assets (Samuelson 1969, Merton 1969, Merton 1971). To explain the non-participation puzzle, the focus of the recent research has been on relaxing or modifying the assumptions of portfolio choice theory. One subset of the literature considers market frictions, such as entry and/or participation costs, as a possible explanation (Haliassos and Bertaut 1995, Bertaut 1998, Haliassos and Michaelides 2003, Paiella 2001, Vissing-Jorgensen 2002, Guiso, Haliassos and Jappelli 2003). The costs comprise a broad set of categories including real costs – brokerage commissions,

Since household wealth surveys have been widely used to study saving and other issues, it is important to examine the reliability of the various survey estimates of wealth. In this paper the authors assess the quality characteristics of the National Longitudinal Survey of Mature Men and the Retirement History Survey, as compared to the 1983 Survey of Consumer Finances. We find that the NLS and especially the RHS underreport wealth and wealth concentration. The underestimates of wealth held in the form of common stock, business equity, and investment real estate equity are substantial. The principal problem lies in underrepresentation of both tails of the wealth and income distributions, with the consequences of underrepresenting the upper tail being especially serious for wealth measurement. We examine several potential reasons for the underrepresentation. In recent years, an extensive literature on life-cycle saving .behavior in the U.S. has developed. The empirical base for this literature consists almost entirely of survey measures of household wealth.' The two most widely used surveys are the National Longitudinal Survey of Mature Men (NLS) and the Retirement History Survey (RHS). The NLS, which began in 1966 with eleven reinterviews between 1967 and 1983, was comprised of males between the ages of 45 and 59 in 1966. A reinterview of surviving NLS respondents is currently underway. The RHS, which began in 1969 with reinterviews every two years until 1979, initially sampled households with heads between the ages of 58 and 63. More recent data sets containing information on wealth include the Survey of Income and Program Participation (SIPP); the Panel Study of Income Dynamics (PSID), where a wealth module was included in the 1984 and 1989 surveys; and the Survey of Consumer Finances (SCF), where a complete wealth survey was conducted in Note: The opinions expressed do not necessarily reflect those of the University of Michigan or the Board of Governors or their respective staffs. The authors would like to thank Arthur Kennickell for invaluable guidance throughout the course of this research, Richard Burkhauser and Robert McIntyre for providing some of the data used in the research, and Daniel Kelley for invaluable research assistance. All errors and omissions are the responsibility of the authors.

Cost-Benefit Analysis (CBA) is an increasingly important form of environmental impact assessment. CBA has long been used for analysing existing air pollution policy proposals and ensuring that the benefits of proposed policies outweigh the costs. However, with the Clean Air Policy package proposed in 2013, the European Commission reversed the conventional approach. Instead of using CBA to calculate economic welfare of proposed emission levels, they used CBA to calculate a precise economic welfare-maximising solution and used the corresponding calculated emission levels as basis for their proposal.Despite the year 2016 adoption of much of the policy package proposal, air pollution will still be problematic in Europe in year 2030. It is expected that most parts of the European Union will experience air pollution levels above those recommended in the new air quality guideline values recently updated by the World Health Organization (WHO). The European Commission is now revising the air quality targets for the European Union, and it reasonable to assume that CBA will once again be influential.At the same time as the influence of CBA has grown with respect to air pollution policy, the literature criticising standard welfare economics and CBA as a concept has increased, albeit in an unfocused way. Given these two opposing trends, it is timely to compile a variety of perspectives and an in-depth discussion as to whether this current use of CBA to support air pollution policies is scientifically robust.This conceptual paper discusses the implicit methodological choices made when using an economic welfare-maximising CBA as an impact assessment tool to set targets in air quality policy proposals. The discussion applies an air pollution CBA-perspective to the existing critique of CBA and welfare economics and adds seminal papers from a well-established body of criticism of economic analysis. Perspectives from the disciplines of behavioural economics, economic psychology, and complexity economics are included and compared in terms of the way in which standard economics represents economic decision-making. This paper is the first to include perspectives from all these disciplines in a discussion of applied air pollution CBAs. The body of criticism is contrasted with theories from science-technology-society studies and an empirical description of the actual process from the production of an impact assessment to a final policy agreement. Potential pathways are discussed, including a discussion of alternatives to the current approach.On balance, minor adjustments, major adjustments (requiring further research), as well as methodological improvements are needed. Desirable minor adjustments include the need to avoid perfect foresight CBAs. It is important to include a range of potential air policy ambitions that are contingent on economic development, climate policy development, and equity preferences. Furthermore, the CBA should be based on a combination of climate and air pollution control options, and not just air pollution control. Appropriate major adjustments include better representation of technology learning, and potentially positive system feedbacks from, for example, electric vehicles. Another major adjustment is the monetisation and inclusion of all known environmental and human health externalities. A final major adjustment is ensuring that CBAs can accommodate existing environmental policy targets as feasible model solutions. The main and important methodological recommendation is that the fundament of CBA methodology should be left as it is, although policy proposal impact assessments should be expanded with analyses made with other methodologies. Such methodologies may adopt other ethical perspectives, based on, for example, egalitarian and hierarchical rationales, when analysing or proposing the ambition levels of air pollution policies.

The analytic and monetary costs and benefits of panel surveys are assessed in light of experiences from the Panel Study of Income Dynamics, an 18-year panel survey on the economic status and behavior of the U.S. population. The analytic benefits of panel are formidable, ranging from description of gross change to various analytic advantages of continous and discrete time modelling. Analytic costs such as the conditioning of responses in subsequent participation or nonresponse bias are possible in panel surveys, but their effects can be minimized with proper data collection procedures and analytic adjustments. Surprisingly, the monetary costs of panel surveys are less than the costs of comparable repeated cross-sectional surveys.

A major development in social science research in the 1970s was the collection and analysis of several large-scale, nationally representative, longitudinal studies. Much has been learned from these studies because they provide more reliable information on changes in the objective and much more detailed point-in-time information than most existing cross-sectional studies and for much larger samples. As a result, they have also been viewed by analysts as better sources of cross-sectional information. However, because of their size and scope, they are seen as expensive ways of collecting data. As the size of the pie for social science funding shrinks, some feel that the value of the smaller-scale research projects foregone by funding panel data collection projects appears too great to warrant panel designs.

This book contains papers prepared for a Workshop on Social Accounting Systems: Critique and Assessment of Prospects held in Washington on March 24-26 1980. The papers focus on describing and evaluating the current state of the art of empirically based approaches to social accounting including both time-based and demographic accounting systems. Papers are included entitled Demographic accounts and the study of social change with applications to the post-World War II United States; The relationship of demographic accounts to national income and product accounts; and The conceptual and empirical strengths and limitations of demographic and time-based accounts.