I. Overview
![](https://assets.pewresearch.org/wp-content/uploads/sites/7/2007/08/2007-wages-01.png)
Foreign-born workers, who now comprise 15% of the U.S. labor force, made notable progress into middle- and high-wage jobs between 1995 and 2005. During this period, many foreign-born Latinos stepped out of the low-wage workforce and headed toward the middle of the wage distribution.
In particular, newly arrived Hispanics proved much less likely to be low-wage workers in 2005 than in 1995. Higher levels of education and a move from jobs in agriculture to construction aided the progress of Latinos in the wage distribution scale. At the same time, there was a large drop in the proportions of workers born in Mexico and Central America who were in the lowest wage class. Among all foreign-born workers, the strongest gains were made by Asians, who significantly boosted their presence in the high-wage workforce.
Despite the gains in wage, foreign-born workers remained more likely than native-born workers to be in low-wage jobs in 2005. And while the pace of growth for foreign-born Latinos in the low-wage workforce was relatively slow, they also accounted for a significant number of new low-wage workers between 1995 and 2005. In part that is because the large initial size of the Latino low-wage workforce translated into high numbers of new low-wage workers even at slower growth rates. At the same time, economic progress for other workers, especially whites and foreign-born Asians, was more rapid than for foreign-born Hispanics.
In contrast to the experiences of foreign-born workers, there was little to no change in the position of native-born groups in the wage distribution between 1995 and 2005. The main exception was the experience of native-born white workers. Their employment growth was largely confined to the high-wage group. Even as their share in the workforce dropped, they added more workers to the high-wage workforce than might have been expected based on demographic trends alone.
These developments came about as two intertwined trends—the growth of the Hispanic population and immigration—have transformed the demography of the U.S. labor market in recent years. The share of Hispanics in the labor force (those with a job or actively looking for a job) stood at 6% in 1980. Their share increased to 9% by 1995 and then to 13% by 2005. At the same time, the share of foreign-born workers in the labor force grew from 7% in 1980 to 11% in 1995 and then to 15% by 2005. These shares will continue to increase in the near future because Latino and foreign-born workers account for the majority of new workers in the economy.
This report uses the prism of the wage distribution to study the integration of foreign-born workers, especially Latinos and Asians, into the U.S. labor market. The analysis addresses key questions about this fast-growing workforce: Are these foreign-born workers crowding into the low-wage segments of the workforce? What are the growth rates for immigrant workers in the middle- and high-income segments of the workforce? How are the newly arrived immigrant workers faring?
Estimates are derived from the Census Bureau’s Current Population Survey (CPS), a monthly survey of approximately 60,000 households. Data on earnings are available for one-quarter of the monthly sample. Twelve monthly samples with earnings data are combined to conduct the analysis on an annual basis. The analysis in this report focuses on changes between 1995 and 2005.1
Both authorized and unauthorized workers are part of the analysis, but this report does not distinguish foreign-born workers by immigration status. For more information on unauthorized workers in the U.S. labor market, see Passel (2006). This report also does not examine the relationship between immigration and wages of native-born workers. That issue has been extensively researched and no definitive answer has emerged. Some economists (Borjas, 2003) have concluded that immigration harms the wages of native-born workers but others (Card, 2001; and Ottaviano and Peri, 2006) find either no effect or a possible beneficial effect on wages. A 2007 report by the Council of Economic Advisers concludes that, on average, native-born workers benefit from immigration.
The study first examines the demographic characteristics of the wage distribution in 2005. The principal characteristics of interest are workers’ nativity, race and ethnicity. That is followed by an analysis of the progress of foreign-born workers in the wage distribution between 1995 and 2005. The last section in the report briefly examines the experience of native-born workers and individual racial and ethnic groups in the same time period.
Assigning Workers to a Wage Group
Workers are assigned to one of five wage groups depending on their rank in the wage distribution. In ascending order, the five wage groups are referred to as low wage, low-middle, middle, high-middle and high. That classification is done once for 1995 and again for 2005.
On average, one-fifth of workers are expected to fall into each of the five wage groups. In other words, about 20% would fall into the low-wage group, 20% into the low-middle group and so on. The lowest earning 20% of the workforce is the low-wage workforce. The highest earning 20% of the workforce is the high-wage group. Workers who fall in the middle 20%, or the third of the five wage classes, are middle-wage workers.
The distribution of individual categories of workers, such as foreign-born or Latinos, may differ from this average. Relatively more or less of a particular group of workers may lie above or below the middle of the wage distribution.
The assignment of a worker to a wage class is based strictly on that worker’s wage relative to the wages of other workers. Thus, the income range that defines the middle of the distribution varies from one period to the next. It should be noted that the terms “low wage,” “middle wage” and “high wage” refer merely to different points of the wage distribution. This report does not attempt to use absolute notions or test alternative definitions of what it means to be a low-, middle- or high-income worker.
In that sense, this report differs from other analyses that focus on the experiences of the middle class (see, for example, Rodriguez, 1996; Bean, Trejo, Capps and Tyler, 2001; and Clark, 2001). Many of those analyses impose an income range that defines the middle class or use an absolute notion of income in conjunction with homeownership to define the middle class. Wolfson (1994) presents a variety of indicators that might be used to measure the spread of the wage distribution or to define the middle of the wage distribution.
The analysis in this report is based on the hourly wage. For workers in the sample who reported only a weekly wage, the hourly wage is estimated as the weekly wage divided by the usual hours worked in a week. Using the hourly wage factors out differences in earnings across workers that arise from differences in hours worked, such as the weekly earnings of full-time versus part-time workers.
The range of earnings for workers in each of the five wage groups is shown in Table 1. The sample for 2005 consists of 122 million workers divided into five groups of approximately 24 million workers each. Low-wage workers in 2005 were earning less than $8.50 per hour and those in the middle were paid $12 to $16.20 per hour. High-wage workers in 2005 earned more than $24.03 per hour.
In 1995 hourly earnings could be estimated for 106 million workers. Thus, there are five wage groups of about 21 million workers each in 1995. Expressed in 2005 dollars, low-wage workers earned less than $7.69 per hour in 1995. Middle-wage workers earned $10.97 to $15.38 per hour, whereas high-wage workers earned more than $22.03 per hour.
The median wage for all workers was $14 per hour in 2005. That was 9.2% higher than the median wage of $12.82 per hour in 1995 (Table 2; all wages are expressed in 2005 dollars). The economy-wide median wage is, by definition, also the median wage of middle-wage workers. Both reflect the midpoint of the wage distribution—half of all workers earn less than the median and half earn more. Therefore, the earnings of middle-wage workers are also seen to have increased 9.2% between 1995 and 2005.
![](https://assets.pewresearch.org/wp-content/uploads/sites/7/2007/08/2007-wages-03.png)
Low-income workers’ median hourly earnings increased from $6.41 in 1995 to $7 per hour in 2005, also a 9.2% increase. Interestingly, the median earning of low-wage workers was exactly half the median earning of middle-wage workers in both years.
High-income workers fared better than average between 1995 and 2005 and pulled away slightly from the rest of the workforce. The median wage of these workers increased from $28.48 in 1995 to $31.73 per hour in 2005, or by 11.4%.
Employment Levels in the Report Understate Actual Employment
The number of workers counted in this report is less than the actual number employed in the labor market. The unincorporated self-employed, workers without pay and those whose earnings are very low (less than $2 per hour) or high (more than $100 per hour) are excluded from the analysis.
In addition, there is some sample loss due to missing data on earnings or usual hours worked. For example, the Bureau of Labor Statistics reported that employment in the U.S. in 2005 was 142 million. However, Table 8 in the report shows 122 million as the total number employed in 2005, or about 85% of the actual level of employment. Thus, employment levels in this report should not be interpreted to represent actual employment in the economy at the point in time in question. The missing observations, representing 15% of employment, would have an effect on the findings in this report only if their characteristics were extremely different from the remainder of employed workers.