-
Published Versus Sample Statistics From The ASM: Implications For The LRD
January 1991
Working Paper Number:
CES-91-01
In principle, the Longitudinal Research Database ( LRD ) which links the establishments in the Annual Survey of Manufactures (ASM) is ideal for examining the dynamics of firm and aggregate behavior. However, the published ASM aggregates are not simply the appropriately weighted sums of establishment data in the LRD . Instead, the published data equal the sum of LRD-based sample estimates and nonsample estimates. The latter reflect adjustments related to sampling error and the imputation of small-establishment data. Differences between the LRD and the ASM raise questions for users of both data sets. For ASM users, time-series variation in the difference indicates potential problems in consistently and reliably estimating the nonsample portion of the ASM. For LRD users, potential sample selection problems arise due to the systematic exclusion of data from small establishments. Microeconomic studies based on the LRD can yield misleading inferences to the extent that small establishments behave differently. Similarly, new economic aggregates constructed from the LRD can yield incorrect estimates of levels and growth rates. This paper documents cross-sectional and time-series differences between ASM and LRD estimates of levels and growth rates of total employment, and compares them with employment estimates provided by Bureau of Labor Statistics and County Business Patterns data. In addition, this paper explores potential adjustments to economic aggregates constructed from the LRD. In particular, the paper reports the results of adjusting LRD-based estimates of gross job creation and destruction to be consistent with net job changes implied by the published ASM figures.
View Full
Paper PDF
-
Returns to Scale in Small and Large U.S. Manufacturing Establishments
September 1990
Working Paper Number:
CES-90-11
The objective of this study is to assess the possibility of differences in the production technologies between large and small establishments in five selected 4-digit SIC manufacturing industries. We particularly focus on estimating returns to scale and then make interferences regarding the efficiency of small businesses relative to large businesses. Using cross-section data for two census years, 1977 and 1982, we estimate a transcendental logarithmic (translog) production model that provides direct estimates of economies of scale parameters for both small and large establishments. Our primary findings are: (i) there are significant differences in the production technologies between small and large establishments; and (ii) based on the scale parameter estimates, small establishments appear to be as efficient as large establishments under normal economic conditions, suggesting that large size is not a necessary condition for efficient production. However, small establishments seem to be unable to maintain constant returns to scale production during economic recession such as that in 1982.
View Full
Paper PDF
-
Estimating A Multivariate Arma Model with Mixed-Frequency Data: An Application to Forecasting U.S. GNP at Monthly Intervals
July 1990
Working Paper Number:
CES-90-05
This paper develops and applies a method for directly estimating a multivariate, autoregressive moving-average (ARMA) model with mixed-frequency, time-series data. Unlike standard, single-frequency methods, the method does not require the data to be transformed to a single frequency (by temporally aggregating higher-frequency data to lower frequencies for interpolating lower-frequency data to higher frequencies) or the model to be restricted by frequency. Subject to computational constraints, the method can handle any number of variable and frequencies. In addition, variable can be treated as temporally aggregated and observed with errors and delays. The key to the method is to view lower-frequency data as periodically missing and to use the missing-data variant of the Kalman filter.
In the application, a bivariate, ARMA model is estimated with monthly observations on total employment and quarterly observations on real GNP, in the U.S., for January 1958 to December 1978. The estimated model is, then, used to compute monthly forecasts of the variables for 1 to 12 months ahead, for January 1979 to December 1988. Compared with GNP forecasts, in particular, for similar periods produced by established econometric and time series models, present GNP forecasts are generally more accurate for 1 to 4 months ahead and about equally or slightly less accurate for 5 to 12 months ahead. The application, thus, shows that the present method is tractable and able to effectively exploit cross-frequency sample information, in ARMA estimate and forecasting, which standard methods cannot exploit at all.
View Full
Paper PDF
-
Gross Job Creation, Gross Job Destruction and Employment Reallocation
June 1990
Working Paper Number:
CES-90-04
This paper measures the heterogeneity of establishment-level employment changes in the U.S. manufacturing sector over the 1972 to 1986 period. Our empirical work exploits a rich data set with approximately 860,000 annual observations on 160,000 manufacturing establishments to calculate rates of gross job creation, gross job destruction, and their sum, gross job reallocation. The central empirical findings are as follows: (1) Based on March-to-March establishment-level employment changes, gross job reallocation averages more than 20% of employment per year. (2) For the manufacturing sector as a whole, March-to-March gross job reallocation varies over time from 17% to 23% of employment per year. (3) Time variation in gross job reallocation is countercyclic-gross job reallocation rates covary negatively with own-sector and manufacturing net employment growth rates. (4) Virtually all of the time variation in gross job reallocation is accounted for by idiosyncratic effects on the establishment growth rate density. Changes in the shape and location of the growth rate density due to aggregate-year effects and sector-year effects cannot explain the observed variation in gross job reallocation. (5) The part of gross job reallocation attributable to idiosyncratic effects fluctuates countercyclically. Combining (3) ' (5), we conclude that the intensity of shifts in the pattern of employment opportunities across establishments exhibits significant countercyclic variation. In preparing the data for this study, we have greatly benefited from the assistance of Robert Bechtold, Timothy Dunne, Cyr Linonis, James Monahan, Al Nucci and other Census Bureau employees at the Center for Economic Studies. We have also benefited from helpful comments by Katherine Abraham, Martin Baily, Fischer Black, Timothy Dunne, David Lilien, Robert McGuckin, Kevin M. Murphy, Larrty Katz, John Wallis, workshop participants at the University of Maryland, the Resource Mobility Session of the Econometric society (Winter 1988 meetings), an NBER conference on Alternative Explanations of Employment Fluctuations, and the NBER's Economic Fluctuations Program Meeting (Summer 1989). Scott Schuh provided excellent research assistance. We gratefully acknowledge the financial assistance of the National Science Foundation (SES-8721031 and SES-8720931), the Hoover Institution, and the Office of Graduate Studies and Research at the University of Maryland. Davis also thanks the National Science Foundation for it's support through a grant to the National Fellows Program at the Hoover Institution. Most of the research for this paper was conducted while Davis was a National Fellow at the Hoover Institution.
View Full
Paper PDF
-
Modelling Technical Progress And Total Factor Productivity: A Plant Level Example
October 1988
Working Paper Number:
CES-88-04
Shifts in the production frontier occur because of changes in technology. A model of how a firm learns to use the new technology, or how it adapts from the first production frontier to the second, is suggested. Two different adaptation paths are embodied in a translog cost function and its attendant cost share equations. The paths are the traditional linear time trend and a learning curve. The model is estimated using establishment level data from a non-regulated industry that underwent a technological shift in the time period covered by the data. The learning curve resulted in more plausible estimates of technical progress and total factor productivity growth patterns. A significant finding is that, at the establishment level, all inputs appear to be substitutes.
View Full
Paper PDF
-
Public Use Microdata: Disclosure And Usefulness
September 1988
Working Paper Number:
CES-88-03
Official statistical agencies such as the Census Bureau and the Bureau of Labor Statistics collect enormous quantities of microdata in statistical surveys. These data are valuable for economic research and market and policy analysis. However, the data cannot be released to the public because of confidentiality commitments to individual respondents. These commitments, coupled with the strong research demand for microdata, have led the agencies to consider various proposals for releasing public use microdata. Most proposals for public use microdata call for the development of surrogate data that disguise the original data. Thus, they involve the addition of measurement errors to the data. In this paper, we examine disclosure issues and explore alternative masking methods for generating panels of useful economic microdata that can be released to researchers. While our analysis applies to all confidential microdata, applications using the Census Bureau's Longitudinal Research Data Base (LRD) are used for illustrative purposes throughout the discussion.
View Full
Paper PDF
-
Long-Run Expectations And Capacity
April 1988
Working Paper Number:
CES-88-01
In this paper, we argue at a general level, that recent economic models of capacity and of its utilization are deficient because they do not adequately take into account firms' long-run expectations about conditions which are pertinent to their investment decisions, i.e., their decisions about altering productive capacity. We argue that the problem with these models is that they rely on the two conventional definitions of capacity which ignore these long-run expectations. Accordingly, we propose a third definition of capacity which incorporates these expectations and, thereby, corrects the problem. Furthermore, we argue that a correct, empirical analysis with the proposed definition -- indeed, any credible analysis of capacity or its utilization -- must take into account the demand for the output produced by the firms being studied. Finally, we apply the definition to clarify the meaning of surveys of capacity and, thus, show how it can be used to improve future surveys of capacity.
View Full
Paper PDF
