Critical Review of CBO's Hourly Wage Distribution Report

Overview and Scope of the CBO 2011 Report

This critical review examines the Congressional Budget Office (CBO) February 2011 publication "Changes in the Distribution of Workers' Hourly Wages Between 1979 and 2009." The review begins with a broad, global description of the paper's major conclusions, then focuses progressively on specific areas of concern, and finally identifies ways the study could be improved. The critical approach moves from a macroscopic overall description down through the major sections of the paper to the details underlying broader conclusions. Particular attention is paid to specific notes where many of the assumptions underlying the paper's conclusions originate. This publication is also compared with similar papers by parallel Congressional and Executive agencies, as well as a few academic sources. Areas for improvement and further research reflect implications and information that are either necessary, useful, or omitted.

A critical review points out both strengths and weaknesses, and there are numerous factors supporting the validity and relevance of this paper. Strengths include that it was published by the Congressional Budget Office — that is, by the U.S. Congress — so the authors have a political stake in publishing accurate information, and their individual biases are at least partially offset through the peer review embodied in that institution. The CBO has also published numerous papers on sufficiently similar topics to permit meaningful comparison. This paper cites prominent but heterodox economists like Paul Krugman and Thomas Lemieux, which improves its credibility compared to some of the examples considered briefly below. Additional strengths are identified alongside weaknesses in the closer analysis that follows.

We must limit our expectations of this document carefully. This February 2011 publication "documents changes in the level and distribution of hourly wages received by workers in the United States between 1979 and 2009" (CBO 2011, unnumbered Preface), and defines some of the major supply and demand factors affecting wages for different skill sets, productivity levels, gender, education levels, and other worker characteristics. The limiting terms "distribution" and "wages" should remain foremost throughout. The paper also considers the role of "labor market institutions" that most affect wages. This all occurs under the CBO's usual disclaimer foreswearing any type of policy recommendation derived from this "objective" and "impartial" study (Preface).

The global conclusions of this paper are that inflation-adjusted wages at the median increased over the period 1979–2009 by approximately 20%; that median wages stood at about $17 at the period's end; that the gap between high and low wages both moved away from the 50th percentile for men and women, for different reasons; and that these changes followed different courses over time. These findings were produced by analyzing wage earnings limited in specific ways, adjusted for inflation under specific constructions, organized into percentiles, and weighted for specific factors analyzed below. The data describe events shaped by market conditions — most importantly technology and productivity innovation, global trade patterns, and immigration — which are treated as separate and distinct from institutional factors such as minimum wage levels and union representation.

Data Set Considerations

The practice of ranking data into different grades and describing the middle and the two tails in reference to the 50th percentile has become so standardized over the past century that criticism of the model itself is beyond the scope of this review. The dataset is large enough; it derives from the U.S. Census and is thus likely to be approximately normal and random. Outliers have been removed by focusing on the tenth and 90th percentiles rather than the absolute tails, which is a sound methodological choice. Downside outliers were further limited by the minimum wage. The use of the median rather than the mean helps overcome distortions in distribution — itself the primary focus of this study — and is likewise robust.

This is roughly the point at which genuine opportunities for controversy emerge. The publication relies on assumptions that begin with the adjustment of data for comparability — so-called "chaining" — both in the indexing for inflation and in the weighting of wages for hours worked. The U.S. Census Current Population Survey data on which this study is based are also not immune from criticism on the grounds of subject reporting errors, as evidenced by sustained criticism from advocates for the Current Employment Survey — a dispute that ultimately reduces to the broader question of administrative versus survey data (Gould 2003). Imputing earnings for missing responses introduces a subordinate potential confound with relatively small but real effects. The most significant problems with the dataset, however, arise through the chaining of wages for hours worked and through specific, questionable decisions made in selecting data for consideration and assigning wage rates for non-wage payroll — or excluding employment classes that probably bear significantly on the validity, relevance, and utility of this report.