Knowledge that Transforms

Robert Moffitt, Walter Nicholson, The Effect of Unemployment Insurance on Unemployment: The Case of Federal Supplemental Benefits, The Review of Economics and Statistics, Vol. 64, No. 1 (Feb., 1982), pp. 1-11

T HE purpose of this paper is to study the small sample performance of tests for causal ordering of time series in the sense of Granger (1969). Versions of three tests are studied: that based directly on Granger's definition of causality and suggested by Sargent (1976); that suggested by Sims (1972); and the modification of Sims suggested by Geweke, Meese, and Dent (1982). Tests for causal orderings of time series have been applied often in recent econometric work. Sims (1972) introduced his version of a causal ordering test to inquire whether money was exogenous (as monetarists might suggest) in the money income relationship. Sargent (1976) used Granger and Sims procedures to test the validity of the natural-rate hypothesis inherent in his model. Salemi (1980) employed the Granger test as a test of specification of a money demand equation in hyperinflation. A goal of our research is to conduct our study with data that closely resemble the types of quarterly time series that arise in applied research. The cost of this approach is a research design for which finitely parameterized versions of the tests are never exactly correct. It is our view that users of these tests are likely to encounter this potential source of bias. Indeed, the problem of truncation (of leading and lagging coefficients in the causality test regressions) arises whenever the vector ARMA representation of the time series studied has, in reality, a nontrivial moving average component. To our knowledge, this feature of our research design has not been used before, and is a major difference between our work and the work of Geweke et al. and Nelson and Schwert (1980). Answers to the following research questions interest us. First, how likely is a user of each version of the test to reach a correct decision regarding the causal ordering of the time series? Second, how accurately does each test procedure recover population values of the test regression coefficients? Third, how sensitive are answers to the first and second questions to sample size, contemporaneous correlation of the exogenous errors, and the strength of the causal interrelationship? Fourth, how important a source of bias is the finite parameterization of the test regressions that is required in small samples? Section II of the paper describes the versions of the tests studied. Section III describes the experimental design. The results of the experiments are in section IV, and conclusions are presented in section V.

Full-information maximum likelihood (FIML) estimates of the simultaneous probit-Tobit (SPT) model suggest that effects of campaign contributions on voting are smaller than single equation probit estimates would indicate. The author has generally unable to conclude that contributions have a significant impact on voting decisions, apparently votes are most often decided on the basis of personal ideology or preferences of constituents. These findings differ markedly from earlier results of economists Gary C. Durden and Jonathan J. Silberman, whose single equation models showed a substantial impact of contributions on voting decisions. Despite the lack of significance according to model SPT, it would not, however, be appropriate to unambiguously conclude that contributions have no effects on voting. For six of eight coefficients the anticipated positive sign resulted and one coefficient remained marginally significant. The article also shows that the lack of significance is attributable not only to smaller coefficient size, but also to larger standard errors. The FIML estimates of the contribution coefficients are not very precise.

N recent years, considerable attention has been devoted to understanding the causes of earnings differentials between blacks and whites. Conventional economic analysis of the issue has been divided into supply and demand factors. Demand side arguments point to differential returns to blacks and whites with the same level of market-valued characteristics as a principal cause of the lower earnings of blacks. Supply side arguments, on the other hand, focus on the quality and quantity of the market-valued characteristics of black compared to white workers. Although this dichotomization is obviously a necessary and useful step in understanding the causes of earnings differentials between blacks and whites, it fails to place sufficient emphasis on the process by which market-valued characteristics are actually obtained. In particular, given the effect of parents' economic status on their children, this approach does not illuminate the intertemporal consequences of racial discrimination for the acquisition of market-valued characteristics. Also, given the potential of the effects of the community of origin on achievement, it may, in addition, obscure many of the negative externalities affecting the acquisition process that result from being part of a discriminated-against group. This paper addresses these problems by examining a recursive model of the effects of socioeconomic background on education and earnings of black and white men ages 23-32. The main difference between this analysis and other efforts in this area is that it examines not only the effects of the socioeconomic status of an individual's parents, but also the effects of characteristics of the individual's community of origin. While other studies have included variables measuring differences due to region and/or size of place of origin, this paper also controls for differences between individuals at the more disaggregated neighborhood level. The findings of the study suggest that neighborhood differences are at least as important as family characteristics in explaining the gaps between black and white achievement and that the omission of neighborhood characteristics results in a misleading picture of the source of background effects.

T HE United States is a country of immigrants and sons and daughters of former immigrants. From colonial times to the 1920s, America had an open door immigration policy. Although the formal policy generally welcomed immigrants, those who preceded the newcomers have never looked kindly on the arrival of new immigrants. Xenophobia has been a common reaction by Americans to large scale inflows. Although ethnocentrism undoubtedly is one reason for such isolationist feelings, fear of economic displacement is another underlying cause.' At the center of the debates over immigration policy has been the fear that the newcomers take jobs away from native Americans. The substitutability of natives for immigrants is critical in evaluating the validity of the displacement fears. A number of theoretical contributions, including Gerking and Mutti (1979), Grossman (1981) and Johnson (1980a) have dealt with the economic effects of international labor inflows. However, no empirical investigation of these issues has been undertaken. The absence of such studies is due, at least in part, to the lack of accurate data on the number of employed illegal immigrants, who comprise a substantial portion of the immigrant stock. While it is currently not possible to develop empirical estimates of the substitutability of illegal immigrants and natives, or even between the most recent wave of legal immigrants and natives, it is nonetheless useful to investigate the relationship in production that has historically existed between immigrants and the domestic labor force. In what follows, I use cross-section data for 1970, the most recent year for which such data are available, to estimate aggregate production relationships. The resulting estimates shed light on the substitutability between the stock of immigrants in the United States at that time, and the native work force. Until more is known about the employment behavior of more recent illegal and legal (especially refugee) immigrants, this is the best that can be accomplished.

An unusually rich data set was tapped to explore the relationship between research and development (R and D) and productivity growth. The most-important finding is that, with disaggregated data, the wrong lag hypothesis is not supported: there is no clear indication that the productivity slump of the 1970s resulted from a decrease in the marginal productivity of R and D. Also important is the evidence of substantial returns to used R and D, i.e., from internal-process work and the purchase of R and D-embodying products, but not (at least in industries with productivity indices based upon physical output or comprehensive price deflators) to the performance of product R and D. Given the fact that three-fourths of all industrial R and D is product-oriented, studies that fail to distinguish between the origination and use of R and D may suffer from appreciable specification errors. Further research using improved R and D and (especially) productivity data is plainly needed, among other things, to clarify the mystery of why estimated R and D-productivity relationships differ for productivity-data subsets of varying quality. 15 references, 3 tables.

Usually, the information a buyer requires in order to obtain the lowest price must be produced at a cost depending on the efficiency with which the buyers gather information. Thus, a suitable price dispersion enables a monopolist to split up the market to permit a more profitable price discrimination. In this paper, necessary and sufficient conditions for the existence of a non-trivial profit maximizing price dispersion are considered. Furthermore, the relationship between statistical properties of the set of consumers' characteristics and this existence problem are studied.

Two basic approaches to using dynamic econometric models to formulate macroeconomic policy are generally in use. These are: invertible transfer function methodology and optimal control methodology. In this paper a technique for determining dynamic growth paths is presented. The result of the application of this technique is the synthesis of the two approaches into a single approach yielding a policy with more desirable characteristics than was previously possible. An example of the use of this synthesis technique using Pindyck's model is presented.

Can we state that at a given Benassy Fixprice Allocation z " there is, say, excess effective demand for a commodity? It turns out that in productive economies there may be ambiguities in the sign of excess effective demand: different effective demand vectors with different signs. may be compatible with z*. We prove: (a) no ambiguity exists if intermediate goods are ruled out and if all firms in the long side of a market perceive binding constraints; (b) in any case one can always select a vector of effective demands yielding minimal sets of buyer's and seller's markets. 1.