Changing Jobs Pays Off

Analysis presented in "Steady Work Pays Off" (refer to the October 1995 issue of Wyoming Labor Force Trends) confirmed that steady workers do in fact earn higher wages than non-steady workers. "In some industries and quarters, steady workers earn more than double the wages of non-steady workers." Steady workers in that study were specifically defined as those individuals who remained with the same employer over the twelve quarter period from first quarter 1992 to fourth quarter 1994.

However, there is another kind of "steady worker" ... the individual who worked each quarter but had one or more job change events during the overall time period. With this type of "steady worker", the question arises, "Do workers' earnings increase after a job change event?" Or, more specifically, "Did individuals between the first quarter of 1992 and the second quarter of 1995 increase their earnings after one employment change event?" This was the subject of a recent study by Research & Planning.

Studying job changers requires starting at nearly the same point as was done when studying steady workers. Two large Department of Employment databases were merged: employee wage information from the Wage Record database was matched against employer data from the Unemployment Insurance (UI) file, which is also called the ES-202 or QUI file.

Wage Record data provides quarterly information on individual workers, organized by Social Security Number (SSN). Wage records contain data on each worker's employment history for up to three jobs. Wage records also contain an employer identification number, which is necessary to match information from the UI file about the employing company's primary industry (such as Mining or Retail Trade). The primary industry is the one which the business is engaged in most of the time.

Thus, two databases were used to obtain information for this study, and employer identification numbers (common to both files), were used to link data from one to the other. Starting with 409,780 UI wage record SSN's (as of October 1995), 28,610 job changers were identified². These job changers were individuals who worked all fourteen quarters (from first quarter 1992 until second quarter 1995), who also had at least one primary job change during the same time period. The primary job is the one at which the individual earns the most money during a quarter (if the individual works more than one job in a quarter). Due to the qualification that job changers worked all fourteen quarters, it also follows that they had a period of unemployment (between primary jobs) of less than three months.

In this study, we define the primary job as the job that pays the highest wage. This does not necessarily imply that the individual considers the job as his or her primary work. For example, a person might be finishing work on a college degree, working part-time in an internship while also working part-time in a convenience store. While the convenience store might pay more, it is probably not what this person would consider his or her primary job. So it is important to note that this study does not give us an idea of which job an individual intends to continue in the future. Similarly, it is certainly conceivable that a person could have two jobs which pay approximately the same wages; in this case, it is possible that the primary job could change back and forth between the two employers each quarter.

These types of situations, and others, clearly affect how the summary data can be interpreted. As in any statistical analysis, such cases must be considered carefully before utilizing the data.

This study found 61.8 percent of the 26,993 job changers³ (16,677) increased their wages after one job change event from their initial jobs. Furthermore, wages increased by an average of $326.50 per quarter, or $1,306.00 over a year (four quarters). Of those 38.2 percent of job changers (10,316) who decreased their wages after an identifiable job change, wages decreased by an average of $269.42 per quarter, or $1,077.67 over a year.

In addition, each individual industry showed a majority of workers increasing wages after the job change event, without exception (see Figure 1). Several industries also showed average yearly wage increases substantially higher than the average increase across all industries (see Figure 2). Job changers in Construction increased their average yearly wage by $1,991; in Agriculture, by $1,915; in Mining, by $1,557; and in Retail Trade, by $1,500. The smallest average yearly wage increase was for job changers in Services ($1,099) and Wholesale Trade ($1,100).

This study also found that most job changers, while they do change the employer they work for, do not change the industry in which they work. For example, among job changers whose primary job was in Manufacturing, 39.4 percent of them found another job in Manufacturing, 18.0 percent found a new job in the Services industry and 13.8 percent found new employment in Retail Trade. In fact, of all those job changers who did change industries, most migrated to Services or Retail Trade. Agriculture was hardly ever chosen as the post-job change industry, probably due to low total employment, low growth and/or low wages in this industry.

Figures 3 through 12 illustrate the different industry patterns created by workers who change jobs. The industry in the title of each figure indicates the industry in which the job changers worked before the job change event and the pie slices in the figure represent the industries in which the job changers worked after the job change event. For example, Figure 6 ("Manufacturing Job Changes") shows which industries individuals initially employed in Manufacturing ended up in after one job change.

There are several questions that could be addressed by a continuation of the research presented here in the future. Demographic data (in particular, age and gender) could be integrated from the Department of Transportation's Driver's License file to determine if subgroups (i.e., job changers who increased wages and job changers who decreased wages) have different demographics. An analysis of more than just one job change could be performed to determine if there is a point at which more job changes result in lower wages. Also, research into the potential relationship between initial income (before the job change) and income after the job change might generate interesting results. For example, do people with low incomes tend to increase or decrease their wages after changing jobs? Financial implications of "steady" work compared to single job changers and multiple job changers is yet another possibility for future study.

Based on the data in this analysis, a single job change results in increased wages for most workers in Wyoming. However, there is also a significant number of people who change jobs and actually lose income. The results also have important implications for job training and employment dollars. Figures 3 through 12 outline job change paths that most people take when faced with an employment choice. Following these paths in training and focusing resources in the appropriate areas would enhance natural employment patterns and the transfer of workers.

Wayne M. Gosar, formerly an Economist with Research & Planning, recently accepted a new position. The job changer analysis presented here was completed prior to his resignation and is presented in its entirety in the publication Tracking Job Changers in Wyoming's Economy.

Gayle C. Edlin, Senior Statistician at Research & Planning and Editor of Trends, condensed the findings of the original report for publication here.

1 This article is based on the report Tracking Job Changers in Wyoming's Economy, by Wayne M. Gosar, which was originally published in July 1996 by Research & Planning. You can receive a complete copy of this report by contacting Research & Planning, P.O. Box 2760, Casper, WY 82602 or calling (307) 473-3807.

2 In fact, 31,768 job changers were originally identified; 3,158 were statistically determined to be outliers and were subsequently excluded from the analysis.

3 There were 1,617 cases missing industry data.