Knowledge that Transforms

ion, he very much doubted that abstraction could provide either understanding of the real world or, by itself, safe guidance for the legislator or statesman. Although Smith failed to absorb some of the valuable analytical contributions of Hume, the physiocrats, and Turgot, he repeatedly amended his major works by bringing into his discussion some neglected variable, some fresh observation of fact, some new objective. He resorted profusely to qualifications, and his models were therefore not rigorous. is arguable, however, by those who, if forced to choose, prefer realism, or at least the pursuit of it, to rigor and elegance of analysis, that both of his major works are on the whole made better by the qualifications he sprinkled in their pages and that he would have made them still better, although still untidier, if he had used even more qualifying adjectives or phrases [28, p. 327]. It is difficult to exaggerate the intellectual stimulus to be derived from reading this and the numerous other provocative bibliographical articles published in the En-

I am grateful to Meghnad Desai, Harry G. Johnson, Marcus H. Miller, Marc Nerlove, R. D. Terrell, J. J. Thomas, and the editors of this journalfor comment and discussion during the preparation of this paper. Although not explicitly referred to in the text, the more theoretical survey article on distributed lags by Griliches [28] also clarified my thoughts in a number of places. Errors of omission and comnmission are, as usual, my own responsibility.

AMPHLETS pose bibliographical problems. The system of scholarly production and consumption is geared easily and smoothly to books and articles, but pamphlets fall between. for example, are reviewed, and articles may be the subject of published comment, reply, rejoinder, final comment and the like. Pamphlets live on after their appearance mainly in footnote citation-which the scholarly apparatus has yet to measure and weigh. An occasional professional periodical like the Economic Journal will list pamphlets by title under Recent Periodicals and Books, and even include a sentence or two of description in 6-point type. Paul Einzig reviewed essays by A. K. Swoboda and F. H. Klopstock on Euro-dollars (Nos. 64 and 65 in Essays) in the Economic Journal for March 1969. But this is rare. For the most part pamphlets are left to drop into the pool of scholarship and swim by themselves if they can. To repair a little of this neglect, this review article addresses the Princeton Essays (plus Studies, Special Papers and Reprints) in International Finance, not to call attention to them, since they are well known, but to celebrate the appearance of the 75th Essay in a series going back to 1943, and perhaps the 40th or 41st in the last 8Jl years under the brilliant editorship of Professor Fritz Machlup. There is far too much to talk about in 3,000 words or less-changing the gold price, seigniorage in international asset creation, crawling pegs, bands, forward exchange, foreign aid, liquidity and the rest. If one went back to the 1950s, one would find more trade, and somewhat less finance, but oil, merchant-marine policies, international cost-sharing and agriculture price policy. There are too many memorable essays; to cite only a few is invidious but inescapable. This exercise is accordingly addressed mainly to bibliographical or bibliophilic issues raised by the series, and only tangentially to substance. Pamphlets are expensive for librarians. They must be accessioned (should the phrase be acceded to?) and catalogued at an average cost of several dollars an item, much higher a rate per page of scholarship than books, because of the lumpiness of the process. This is because they are not indexed with joumal articles. In a curriculum vitae pamphlets belong under articles, in the library apparatus under books. But not all retrieval apparatus will cope with pamphlets on the same basis as books: the AEA Periodical Index will do well to include the standard series of pamphlets, such as the Essays, Studies, Special Papers in its special volume for lost literature-an index of articles in Festschriften, symposia, conference proceedings. The library problem aside, what of the scholar? Should he file his Essays alphabeti cally by author in the vertical files housing reprints in his outer office, or shelve them in the inner sanctum with books? A survey

This paper is part of a Ph.D. dissertation submitted by the author to the Graduate College, University of Iowa, and was partially supported by the National Science Foundation Grant GS-1491. The author acknowledges the guidance and encouragement received from Professor S. Y. Wu.

T HE geometric distributed lag model developed by Koyck [8] and Nerlove [11] has been widely used in many empirical studies. The appropriate estimation method for parameters of these models is determined by the true probability distribution of random errors. If the random errors follow a first-order Markov process, Klein [7] has shown that the method of weighted regressions yields maximum likelihood estimators.' However, Klein's method requires prior knowledge of the true serial correlation of random errors. In the absence of such prior knowledge, one can resort to several alternative estimation methods.2 In this paper, I establish a bracketing rule applicable to a subset of the admissible probability distributions of random errors. Consider two special cases of Klein's method in which the true serial correlation p is (i) equal to zero and (ii) equal to the coefficient of the lagged dependent variable (1 X). The former case implies an orthogonal regression while the latter is equivalent to ordinary least squares (OLS). The orthogonal and OLS regressions yield two sets of parameter estimates lying on either side of the maximum likelihood parameter estimates provided that the true serial correlation lies in the interval 0 < p < 1 L, the OLS estimate of the population parameter 1 X. This basic bracketing theorem can be shown to hold even for fixed sample size. Moreover, the width of the interval bracketing the maximum likelihood parameter estimates is narrower, the larger is the partial correlation with the lagged dependent variable. This last result leads to an important implication. If a geometric distributed lag constitutes the correct specification of economic behavior, the dependent variable should be highly correlated with the lagged dependent variable. In this event, the discrepancy between OLS and orthogonal parameter estimates (which bracket the maximum likelihood estimates) will be small. Thus, the bias due to least squares is negligibly small provided that the true serial correlation lies in the interval 0 < p < 1 X. A simple geometric distributed lag model is described by a system of two structural equations. Yt a + 8Zt* + Ut + (1)