Printer Friendly

Labor-market data: supplementary sources.

Labor-market data: supplementary sources

In the past, private organizations and State government agencies attempted to fill some of the statistical gaps left by the Bureau of Labor Statistics. Current evidence suggests that the same tendency still exists: if there is a market for data, some organization often steps in to provide them, either for reasons of public relations, or as a direct item for sale. In addition, State statistical agencies will provide information felt to be useful within their jurisdictions.

To illustrate these sources--and their pitfalls--two areas are discussed below: salary intention surveys and State industrial relations data.1

Salary intention surveys

Although it is possible to collect data on expectations and intentions (as the Commerce Department does with regard to plant and equipment expenditures), BLS has not collected data on planned pay adjustments. Some information is of potential use to pay setters and to economic forecasters, and some management consulting firms do survey such information.

As an example, data are collected by Hewitt Associates on pay adjustments planned and under way for salaried employees. We compared the Hewitt figures with realized wage adjustments for white-collar workers taken from the Employment Cost Index. It appears that surveyed personnel managers at first underestimated the degree of wage disinflation occurring in the early 1980's, but then stabilized their expectations in line with actual results. Thus, the Hewitt data provide insight into the shift of wage norms that developed during the economic downturn of 1980-82.

Unfortunately, use of salary intention surveys is hindered by the misunderstanding common among personnel managers concerning the cost of "merit' increases. Particularly among nonunion employers, there is often a confusion between the gross and net effects of merit pay awards. In a steady-state situation, a properly operated merit system (in which across-the-board adjustments are segregated from merit awards) should not raise average pay.2 Yet respondents to the Hewitt and other surveys seem to include gross merit awards in their estimates, thus biasing up the figures by roughly 1 to 2 percent. These upward-biased estimates are then cited, giving a misleading indication of likely wage trends.3 The merit problem illustrates the more general methodological weakness sometimes associated with private data suppliers.

State industrial relations data

Although some State labor statistics agencies predate the BLS, they have had a much less visible role collecting data in modern times. Often, data available from State agencies are derived from BLS or Census series. But in some States, the agency collects industrial relations data on its own. For example, the California Department of Industrial Relations puts out data on union wage settlements and union membership by industry and region.

It is unlikely, however, that State agencies will quickly fill gaps left by the reduction of BLS data collection. For example, eight States were reported to have issued union membership data during 1984, according to the Statistical Reference Index. But closer inspection reveals that all but three (California, South Carolina, and Wyoming) are still reproducing the now-discontinued BLS series from 1978 or 1980. States which collected their own membership prior to the BLS discontinuation continue to do so; the others have not been motivated to undertake the effort.

TO THE EXTENT that a market or a public relations value is perceived for collecting labor market data, private sector organizations often undertake the task. However, general availability of such data for research purposes can be a problem. And problems of methodology (sampling, precise definitions, technical explanations) are less likely to concern private suppliers than BLS. Private organizations have less authority than a government agency in requesting cooperation with surveys; potential respondents may have concerns about confidentiality and the use to which data will be put; and the users themselves may be less sophisticated than statistical technicians about methodological issues. These factors suggest that private collection--while playing a useful role in data provision--is really a complement to, rather than a substitute for, Federally collected data.

State government statistical bureaus do have a level of authority not found in the private sector. But they have tended to become reliant on breakdowns from Federal data sources for much of their output. And the statistical output which State agencies produce is largely applicable only within their borders.

1 References to non-BLS data sources can be found in Margaret A. Chaplan, Labor Statistics: The BLS and Beyond (Champaign, University of Illinois, 1984), Reprint 322; and Katherine I. Bagin and Kevin P. Barry, Unexpected Sources of Information in Industrial Relations: A Current Awareness Approach (Princeton, NJ, Princeton University, Industrial Relations Section, 1984).

2 Imagine a formal progression plan with a series of defined merit steps. As long as the proportion of employees at each step is constant, the average wage will not change. In the steady state, the number of employees retiring from the top will be offset by those entering from the bottom. Thus, although existing workers may be receiving large merit increases (depending on the gap between steps), the average wage will remain constant, Confusion over this issue is rampant because managers are often given "merit budgets' as a control device to prevent them from finding "too many' employees to be especially meritorious. These merit budgets often are based on gross cost or may include what amounts to across-the-board money designed to raise the average wage. See Arnold R. Weber and Daniel J.B. Mitchell, The Pay Board's Progress: Wage Controls in Phase II (Washington, The Brookings Institution, 1978), pp. 89-93.

3 Hewitt's questionnaire asks respondents to calculate a salary structure increase based on the movement of the midpoint of salary ranges and an average base salary increase. The former is essentially a rate range adjustment and should be free of any merit system "taint.' The latter, defined as the increase in the average wage per employee, ought to include only the net cost of merit (which in the steady state should be zero). Yet, it is typically 1 to 2 percent higher than the former, suggesting respondents are using a gross cost of merit in their calculations. (When Hewitt asked its respondents in late 1984 whether they were following the precise instructions of the questionnaire, 70 percent said "yes,' suggesting that the problem is based on inadvertent misunderstanding of the impact of merit pay.) Unfortunately, it is the base salary increase (and similar estimates from other surveys) that tends to be reported. (See, for example, Audrey Freedman and others, Labor Outlook 1985 (New York, The Conference Board, 1984), p. 9.
COPYRIGHT 1986 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1986 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:sources of statistics for wages and industrial relations
Author:Jacoby, Sanford M.; Mitchell, Daniel J.B.
Publication:Monthly Labor Review
Date:Jun 1, 1986
Previous Article:BLS and Alice Hamilton: pioneers in industrial health.
Next Article:Airline deregulation and labor relations.

Related Articles
Gaps in monitoring wages and industrial relations.
Needed: an interdisciplinary approach to labor markets and wage determination.
Family-related benefits in the workplace.
The Employee Turnover and Job Openings Survey.
Differences in state unemployment rates: the role of labor and product market structural shifts.
Can occupational labor shortages be identified using available data?
Manufacturing employment in China: the scale of manufacturing employment in China dwarfs the numbers of manufacturing workers in other countries;...
Manufacturing earnings and compensation in China: on the basis of published earnings data, estimated compensation ratios, and estimated hours,...
Employment, unemployment, and wages.
Notes on current labor statistics.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters