On the asymptotic optimality of the LIML estimator with possibly many instruments

We consider the estimation of the coefficients of a linear structural equation in a simultaneous equation system when there are many instrumental variables. We derive some asymptotic properties of the limited information maximum likelihood (LIML) estimator when the number of instruments is large; some of these results are new as well as old, and we relate them to results in some recent studies. We have found that the variance of the limiting distribution of the LIML estimator and its modifications often attain the asymptotic lower bound when the number of instruments is large and the disturbance terms are not necessarily normally distributed, that is, for the micro-econometric models of some cases recently called many instruments and many weak instruments.     

TIME SERIES ANALYSIS OF ACCOUNTING POLICY CHOICE: REPLY

Taylor (1992) argues that accounting policy choices in any year are not assessed independently of policy choices adopted in previous years. This independence assumption is a maintained hypothesis of Anderson and Zimmer (1992). Although Taylor acknowledges that our results remain robust to his suggested adjustments for non-independence, we argue in this reply that such adjustments are unnecessary because temporal independence of accounting policy choices is consistent with the implications of costly contracting between the firm and its claimholders. We develop the arguments that (1) accounting choices are temporally independent and our research design is therefore, appropriate (2) to the extent that there is an independence problem of the type proposed by Taylor (1992) it also applies to cross-sectional studies of accounting policy choice.     

Development of agriculture in Ethiopia since the 1975 land reform

A stated objective of the Ethiopian Government since it came to power in 1975 has been to increase the production of crops for both domestic consumption and export. Consequently, a number of plans and programs have been proposed and implemented to bring about such development in Ethiopian agriculture. This paper mainly focuses 011 a macro review of the general course of Ethiopian agricultural development since the land reform of 1975. The food production in Ethiopia has lagged behind the population growth as evidenced by the gap between the growth rates of population (3%) and food production (1.8%) over the period from 1970 to 1983. Recently, this situation has, perhaps, further deteriorated. Seventy-one percent of the total land mass is estimated to be suitable for agriculture, but only about 19% is cultivated. This suggests an underutilisation of land. Moreover, Ethiopia has a persistent subsistence production problem in agriculture. Agricultural research and education which are essential to development have not been given due attention by the policy makers. As a result, the agricultural technologies used by farmers have changed little in this century. Moreover, small-scale farmers, although still the most potent economic force in the country, have not been given the incentives necessary to expand production, and currently receive less than adequate support from the central government. Some 85%, 50%, and 79% of total agricultural credit, fertiliser and improved seeds respectively go directly to socialist enterprises (state farms and producers' cooperatives) which, up to 1985, have jointly accounted for only 5% of the total cultivated land and 4% of the national crop production. The favouring of large-scale and capital-intensive state-owned farming enterprises with credit, subsidies, and fiscal incentives, while neglecting smallholders, has contributed to the stagnation of agricultural production in Ethiopia. Overall, the analysis indicates that there has been no significant development of agriculture in Ethiopia following the 1975 land reform.     

An appraisal of the Box-Jenkins approach to univariate time series analysis

Summary Some comments are made on the current methodology associated with theBox-Jenkins approach. The three strokes of Identification, Estimation and Verification are considered; and a final stage, that of Interpretation, mentioned. A discussion of whether the approach is practicable is also given.     

If Time Is Money Is It a Common Currency? Time in Anglo,Asian, and Latin Cultures

ABSTRACT

Social time affects the timing and frequency of purchases, the hours consumers patronize businesses, and how long consumers expect products to last. Understanding how culturally based social time attitudes differ will help marketers tailor their offerings and messages to consumers whose concepts of time differ. This exploratory study combined related streams of time literature to examine constructs of social time differences across cultures. Drawing upon a sample of 1377 respondents from six countries representing three cultural groups (Anglo, Latin, and Asian), the study explores the similarity and differences of cultural social time differences among these three groups.     

INCOME TAX DATA AND FARM FINANCIAL STATISTICS

Tracing the income patterns of individual farm operators whose major source of income is generally derived from farming indicates that off-farm income is becoming an increasingly important income component. The relative importance of off-farm income has nearly quadrupled during the last 20 years, rising from about 10 percent of total income to about 37 percent of total income. Most of this off-farm income can be traced to wages and salaries. These fractions are heavily dependent upon relative income levels. In 1970, very low income farmers posted an aggregate net farm loss whereas the $15,000 - $20,000 class secured about three-quarters of their total income from farm sources. In this same year, wages and salaries were generally the prime off-farm income source for farmers with an assessed income of less than $20,000. Dividends and interest were more important to farmers with an income in excess of $20,000. These differences are faithfully reflected in the regional compilations. As a percent of total income, off-farm income ranged from 41 percent in Ontario to about 29 percent in Quebec. The composition of off-farm sources also varies considerably between regions. In 1970 the income position of taxable unincorporated farm operators (who relied on the farm for their major source of income) remained relatively unfavorable. These farmers still had one of the lowest average incomes of any major occupational class in Canada, with an income distribution which was relatively equally distributed between income classes and not unlike that of the national average. These assessed income statistics, of course, are not necessarily indicative of the welfare position of the respective occupational classes. The calculations conducted are simply illustrative of how income tax statistics can be utilized to facilitate our understanding of the income structure in primary agriculture today. The exercise simply underlines the observation that: The fact that there are conceptual differences between income tax data and other data sources should not detract from the usefulness of the income tax data …. Income tax statistics can stand as an independent data source for the analysis of a variety of issues [13]. For many purposes the tax definition of a farm operator may be more operational than the concept of a farmer as defined in the Census of Agriculture. We might profit considerably from further analyses which attempt to disaggregate Census farms into more homogeneous farm-types using the tax statistics available.