Journal of Economic and Social Measurement - Volume 24, issue 2

Authors: Renfro, Charles G.

Article Type: Research Article

Abstract: This paper introduces the latest report of the Seminar on Model Comparisons, being a set of articles that describe subsequent effects on representative US macroeconometric models of the most recent benchmark revisions in the US National Income and Product Accounts. …These accounts state the definitional relationships between measured macroeconomic variables. They also provide an evolving view of the characteristics of the US economy, this evolution occuring as the result of occasional changes in the accounting constructs. In some cases, the changes simply reflect the effect of differences in the availability of the underlying primary measurements. However, in other cases, they are made in order to bring the accounting concepts into better conformity with economic concepts, or for the sake of harmonizing the US accounts with those of other countries. In all cases, such changes may nevertheless affect the particular use of these accounts, implying the need for users to make a careful evaluation of the implications. Show more

DOI: 10.3233/JEM-1998-0146

Citation: Journal of Economic and Social Measurement, vol. 24, no. 2, pp. 63-82, 1998

Authors: Lasky, Mark J.

Article Type: Research Article

Abstract: In January 1996, the Bureau of Economic Analysis switched from fixed-weighted GDP to chain-type GDP as its featured measure of real output, because fixed weights are appropriate only when the relative price structure of the economy does not change over …time. This paper shows that the better microeconomic properties of the chain-type measures also help the simulation properties of a macroeconomic forecasting model. In particular, Laspeyres-based models often violate the assumption that marginal revenue equals marginal cost. This can produce unrealistic income and multiplier responses, which are eliminated by using chain-type output. Show more

DOI: 10.3233/JEM-1998-0145

Citation: Journal of Economic and Social Measurement, vol. 24, no. 2, pp. 83-108, 1998

Authors: Witte, Willard E. | Green, R. Jeffery

Article Type: Research Article

Abstract: At the end of 1995 the Bureau of Economic Analysis introduced chain weighted Fisher ideal indexes as the primary measures both for real GDP and its components and for the price data associated with the output measures. In response to …these changes macroeconomic modelers were forced to rework the affected sections of their models. This paper describes the accommodation of the Indiana University Econometric Model of the U.S. (Indiana EMUS) to these changes. We briefly describe the overall structure of the Indiana EMUS and outline our overall strategy for dealing with the chain-weighted data. We chose to focus the model on output data stated in terms of chained 1992 dollars, rather than the index numbers themselves. We decided to utilize implicit deflators rather than the chain weighted indexes as our price variables. The main focus of our discussion is the price determination portion of the model, and some of the issues, in that regard, we have confronted in adjusting to the new data regime. Show more

DOI: 10.3233/JEM-1998-0144

Citation: Journal of Economic and Social Measurement, vol. 24, no. 2, pp. 109-122, 1998

Authors: Varvares, Chris | Prakken, Joel | Lisa Guirl, Lisa

Article Type: Research Article

Abstract: The switch to chain-type GDP eliminated the substitution bias problem that plagued the fixed-weight aggregative data macroeconomists have been accustomed to working with. In this paper we review the essential differences between fixed-weight and chain-type measures of output, as well …as the source of the substitution bias. The elimination of the substitution bias makes it worth facing the considerable difficulties occasioned by the switch. Therefore modelers must be prepared to take the necessary steps to re-structure their models to accommodate the new regime. The major tasks in this process are: 1) Re-estimate the behavioral relationships. Generally, where aggregates are employed on the left or right hand side of equations, eliminating the bias inherent in the fixed-weight data will improve the regressions. However, one must pay special attention to issues of scaling, especially where coefficients have familiar interpretations. For this reason, it makes sense to use dollar-denominated quantity indices where possible. 2) If more than one aggregation scheme is to be employed, and it may well be a good idea, modelers must be prepared to re-structure their models to remove any aggregate series from the simultaneous block of the model and reduce aggregation to a table-writing problem by making it totally post simultaneous. 3) Be prepared to add possibly thousands of lines of additional code to accommodate the new aggregation method. Even then, it will probably be necessary to incorporate aggregation residuals and PxQ residuals to insure that the model can replicate the official figures. Finally, even today, more than two years beyond the switch to chain-type GDP in the US, there is a gross lack of appreciation for the problems related to working with chain-type GDP. To the extent that groups such as ours have handled the ugly details and incorporated into our models the needed modifications, perhaps we have made it a kind of black box. Indeed, this is the case in our system. Users can be totally ignorant of the details of chain-type GDP and still generate results that observe the rules of its construction. Show more

DOI: 10.3233/JEM-1998-0149

Citation: Journal of Economic and Social Measurement, vol. 24, no. 2, pp. 123-142, 1998

Authors: Bachman, Daniel | Jaquette, Peter | Karl, Kurt | Rocco, Pasquale

Article Type: Research Article

Abstract: We describe the impact of the new GDP methodology on WEFA's Quarterly model. First we briefly describe WEFA's Mark 11 Quarterly model and its uses, which create certain restrictions on the model specification. We then discuss the question of the …possible impact of the new methodology on estimated elasticities and model multipliers. Since we adopted a new econometric methodology with the new model, the elasticities cannot be usefully compared. Finally, we describe our solution to the ``adding up'' problem created by BEA's complex Fisher Index number approximation. The new WEFA model uses a chained Laysperes index formula to approximate BEA's complete formula. Show more

DOI: 10.3233/JEM-1998-0147

Citation: Journal of Economic and Social Measurement, vol. 24, no. 2, pp. 143-154, 1998