May 29, 1998
New Findings from Oregon Suggest Minimum Wage Increases
Can Boost Wages for Welfare Recipients Moving to Work
by Ed Lazere
Introduction and Summary
Table of Contents
Debates over raising the minimum wage often focus on the issue of who will benefit from lifting the wage floor. Some claim that increasing the minimum wage does little for low-income families, because minimum wage earners frequently are teenagers or secondary workers in middle-class families. While the benefits of minimum wage increases are not targeted solely on low-income families, many of the workers whose earnings rise as a result of such increases are in low-income families. In addition, the large majority of minimum wage workers are adults, and many are the primary breadwinner in their family.
New evidence from Oregon suggests that minimum wage increases can have a significant effect on parents who leave welfare for work. As a result of a successful state ballot initiative, the Oregon minimum wage rose from $4.75 an hour to $5.50 an hour in January 1997 and then to $6.00 an hour in January 1998.(1) Data from Oregon's welfare agency show that the earnings of parents who moved from welfare to work were boosted as a result of these increases.
- The average starting wage of Oregon parents leaving welfare fell five percent, adjusting for inflation, during the three years prior to the January 1997 increase in the state minimum wage.
- This trend reversed itself immediately after the increase. The average starting wage jumped from $6.15 an hour in the last quarter before the 1997 minimum wage increase took effect to $6.43 an hour in the first quarter following the increase. By the fourth quarter of 1997, the average starting wage had reached $6.65 an hour, an increase of more than five percent, adjusting for inflation, over the same quarter of the prior year.
These wage increases are unlikely to have occurred simply as a result of the state's strong economy. Economic growth in Oregon was solid prior to the state's minimum wage increase in 1997, but real wages for welfare recipients still declined during that period.
- The average starting wage for welfare recipients rose again following the second increase in the minimum wage. In the first quarter of 1998, welfare recipients who found work earned an average of $6.91 an hour, up substantially from the end of 1997. Overall, the average wage for these workers is now 76 cents an hour higher than before the state minimum wage increase took effect.
The minimum wage increase in Oregon boosted the earnings both of welfare recipients who found jobs at the minimum wage and of many who found jobs paying slightly above the minimum wage. For example, among welfare recipients who found full-time jobs, the proportion earning more than $6 an hour rose from roughly half in 1996 to two-thirds in 1997.
This analysis also examines whether the increase in Oregon's minimum wage had a negative overall effect on employment opportunities of welfare recipients. While no systematic study of this issue has been conducted, the available evidence does not suggest a negative change in employment opportunities. The share of welfare recipients finding work rose modestly in 1997 following the increase in the minimum wage. In addition, employment growth in retail trade, the industry most likely to be affected by a minimum wage increase, was positive in 1997 and followed the same pattern as overall employment growth in Oregon. Both overall employment and retail employment rose in 1997, although at a somewhat slower rate than in 1996. The change in employment growth between 1996 and 1997 reflects a modest general slowdown in the state's rate of economic growth, not the increase in the minimum wage. If the minimum wage increase had reduced job growth significantly, it is likely that the trend in retail trade employment would have been significantly worse than the trend in overall employment.
The Oregon findings have implications for the debate over raising the federal minimum wage. The Oregon minimum wage hikes have raised the state's minimum wage from a level equal to the federal minimum wage to a level somewhat above the current federal minimum wage. (In late 1996, the federal minimum wage rose to $4.75 an hour, which made it equal to the 1996 Oregon minimum wage. The federal minimum wage then rose to $5.15 an hour in September 1997, below the 1997 Oregon minimum wage of $5.50 an hour at that time.)
In addition, academic studies and recent findings from state welfare-to-work efforts indicate that adults who leave welfare for work typically have very low earnings, often at the minimum wage or just slightly above it. As a result of these low earnings, many families in which parents leave welfare for work continue to live in poverty. Thus, the Oregon experience suggests that an increase in the national minimum wage to a level modestly above $5.15 an hour would be likely to have positive effects nationwide on the wages of adults leaving welfare for work. For the same reasons, the Oregon findings also suggest that increases in state minimum wages could boost the earnings of welfare recipients when they find work.
The Wages of Oregon Welfare Recipients Who Found Work Before and After the State's Minimum Wage Increase
The Adult and Family Services Division of the Oregon Department of Human Resources is the agency responsible for the state's welfare to work programs. The agency regularly collects information on the earnings of welfare recipients who move into the labor force.(2) The Oregon data show a distinct change in the wage trend among welfare recipients who found work in 1997, following the increase in the state's minimum wage. Between 1993 and 1996 the state's minimum wage remained at $4.75 an hour. During that period:
- The average hourly starting wage of welfare recipients who found jobs including both part-time and full-time jobs went from just under six dollars an hour in 1993 to just above that level in 1996.
- When adjusted for inflation, the average starting wage of welfare recipients who found work was five percent lower in 1996 than the average in 1993. (See figure and table.)
These trends changed markedly immediately following the increase in Oregon's minimum wage on January 1, 1997. The average wage of Oregon's welfare recipients moving to work began to rise noticeably after the first boost in the state's minimum wage, and again after the second boost.
Between the last quarter of 1996 and the first quarter of 1997, the average wage for this group rose 28 cents an hour. This was the largest quarter-to-quarter change in the entire period from 1993 through 1997.
- By the last quarter of 1997, the average wage stood at $6.65 an hour, or 50 cents higher than in the similar period of 1996. Adjusting for inflation, this represented a one-year wage increase of 5.4 percent.
- The Oregon minimum wage rose to $6.00 an hour in January 1, 1998. Early data following this increase suggest that it also has had a significant positive impact on the wages of welfare recipients who find work. The average wage for newly employed welfare recipients reached $6.91 in the first quarter of 1998.
Average Starting Wages for Oregon Welfare
Recipients who Found Work 1993 to Early 1998
Average Starting Wage Before Minimum Wage Hike* In Nominal
1993 $5.89 $6.54 1994 5.95 6.44 1995 6.01 6.33 1996 6.08 6.22
4th Quarter, 1996
6.15 6.27 After 1997 Minimum Wage Hike Average 6.52 6.52
6.65 6.61 After 1998 Minimum Wage Hike
6.91 6.85 * Annual figures reflect averages of quarterly figures for the relevant year.
Source: Oregon Adult and Family Services Division, Department of Human Resources
The minimum wage increase appears to have affected a substantial proportion of Oregon's welfare recipients who found work in 1997. In the last quarter of 1997, for example, nearly one-third of the welfare recipients who found work including half of those who found part-time work earned the state's new minimum wage of $5.50 an hour. This substantial clustering of wages at the new minimum wage level strongly suggests that many of these workers would have earned less than $5.50 without the increase in the state's wage floor.(3)
In addition to raising the earnings of minimum wage workers directly, the Oregon minimum wage increase also appears to have raised the wages of workers with earnings modestly above the new minimum wage level. During Oregon's 1996 fiscal year (from July 1995 to June 1996), roughly half 48 percent of the welfare recipients who found full-time work earned more than six dollars an hour. By the last quarter of calendar year 1997, more than two-thirds 68 percent of welfare recipients who found full-time work had earnings this high. Most of this increase occurred in the share of workers earning between six dollars and eight dollars an hour, or somewhat above the new minimum wage level of $5.50 an hour. Minimum wage increases typically have such a "ripple effect," which may reflect attempts by employers to maintain pay differences between minimum wage workers and those with slightly higher earnings when minimum wages rise.
The Oregon minimum wage increase thus affected both welfare recipients who found work at the minimum wage and those who found jobs paying slightly above the minimum wage, which together represent a sizable share of Oregon welfare recipients who move into jobs. As a result of the broad effect, the average increase in the wages of all welfare recipients who moved into the workforce 50 cents an hour from the end of 1996 to the end of 1997 equaled a substantial fraction of the 75 cent increase in the state's hourly minimum wage.
It should be noted that the wage increases revealed by the Oregon data could reflect in part the state's growing economy and its impact on wages. For three reasons, however, the economy appears unlikely to be the primary cause. First, Oregon experienced solid economic growth prior to the 1997 minimum wage hike. Per capita income of Oregon residents rose seven percent between 1993 and 1996, adjusting for inflation. This was greater than the five percent growth in inflation-adjusted per capita income for the nation as a whole during this period. Yet despite this strong growth, the starting wages of Oregon welfare recipients who found work fell in inflation-adjusted terms during this period.
Second, while the state's continued economic expansion could have resulted in a growing shortage of available workers, which would have placed upward pressure on wages, this does not appear to be the case. The state's unemployment rate fell from 7.3 percent in 1993 to 4.8 percent in 1995, but then rose to 5.9 percent in 1996. The rate then remained steady at 5.8 percent in 1997. If the state were experiencing a significant shortage of workers, it would be reflected in a declining unemployment rate in recent years.
Third, the 1997 wage increases among Oregon welfare recipients stand out when compared with wage data for the nation as a whole. While national-level data on the wages of adults who move from welfare to work are not available, the Oregon findings can be compared with the wage trend nationally for a similar group of low-wage working women.(4) The comparison shows that prior to the Oregon minimum wage increase, the wage trend was worse among Oregon welfare recipients who found work than among low-wage working women nationally. Following the 1997 state minimum wage increase, however, wage growth for Oregon welfare recipients was greater than among low-wage women nationwide.
- Between 1993 and 1996, the average wage among welfare recipients in Oregon fell nearly five percent, adjusting for inflation, compared with a one percent decline in the real wages of the typical low-wage female worker nationally.
- In 1997, the average hourly wage for Oregon welfare recipients moving into jobs rose 5.4 percent, also adjusting for inflation, while the typical low-wage female worker experienced a one percent increase in real wages.(5)
Assessing the Employment Effects
One of the major concerns raised about raising the minimum wage is that it can reduce employment prospects for low-skilled workers, under the assumption that some employers reduce the number of employees or the hours they work to offset the increase in wage costs per employee. In recent years, some have argued that the employment prospects of welfare recipients in particular would be adversely affected by an increase in the minimum wage. At the state level, some also argue that an increase in the state minimum wage would result in the migration of jobs to neighboring states.
While these concerns are important and must be considered, the weight of recent research findings on the minimum wage suggests that moderate increases from current minimum wage levels would not have adverse employment effects. No systematic studies of the employment impact of the Oregon minimum wage increase have been conducted to date. Two available measures, however the proportion of welfare recipients moving to work and employment growth in retail trade, the industry most affected by the minimum wage do not suggest that the state's minimum wage increase had a significant adverse effect on employment opportunities.
Employment Trends Among Welfare Recipients
The Oregon Adult and Family Services Division reports that the share of welfare recipients who found work in 1997 7.3 percent in an average quarter was slightly higher than in 1996 before the state's minimum wage was increased. In 1996, an average of 6.4 percent of welfare recipients found work in an average quarter. At the same time, the number of welfare recipients who found jobs declined during this period, from an average of 5,850 a quarter in 1996 to an average of 4,875 per quarter in 1997. Because the state's total welfare caseload fell by a greater magnitude during this period, the percentage of all recipients who found work rose modestly.
These findings support an interpretation that job opportunities for Oregon welfare recipients did not worsen in 1997. It is reasonable to expect that as the total number of welfare recipients declines, the number finding jobs would also decline, simply because the number of welfare recipients who potentially could find work would be smaller. If job opportunities for welfare recipients had worsened significantly, both the number and proportion of welfare recipients finding work would likely drop, as more welfare recipients stayed on the rolls without a job.
At the same time, welfare caseloads may fall for a variety of reasons, including welfare policy changes. If caseloads decline significantly among groups of families that would not have been likely to find work had they remained on welfare, the number of welfare recipients finding jobs would not necessarily drop significantly. This could occur, for example, if changes in welfare programs made assistance less attractive or more difficult to obtain or if some families chose to leave welfare or not to begin receiving welfare to avoid using up time-limited assistance. (Such families may be able to get support from friends or relatives.) It is important to note that welfare caseloads have declined faster in Oregon in recent years than in almost every other state. While the state's caseload decline is partly the result of a strong economy, it clearly reflects policy changes as well.
Because welfare caseloads in Oregon have declined dramatically, and because the characteristics of the state's welfare recipients may have changed significantly, it is difficult to use the data on the proportion of welfare recipients finding work to assess with certainty whether job opportunities for welfare recipients have changed significantly. Nevertheless, the data do not indicate that finding a job became harder for welfare recipients in 1997.
Retail Trade Employment
Retail trade is the industry with the largest proportion of employees earning at or near the minimum wage. It is likely that employment growth in retail trade would be weaker than employment trends in other industries if an increase in the minimum wage had led businesses to reduce the number of employees. This did not happen in Oregon in 1997 following the increase in the state's minimum wage.
Table II shows the growth rate in retail trade employment and in all other industries in Oregon since 1995. There are two things to note about this table. First, each year during this period, retail trade employment rose at a slower pace than the rest of the labor force. Because employment in retail trade was growing more slowly than employment in other industries before the state's minimum wage was raised, the lower growth rate of retail employment in 1997 reflects a continuation of a trend and does not appear, at first blush, to be related to the minimum wage increase.
Oregon Employment Growth Rates, 1995-1997
1995 1996 1997 Total non-farm trade 4.1% 4.0% 3.4% Retail trade 3.4% 3.1% 2.8% Total less retail trade 4.2% 4.2% 3.5% Source: U.S. Bureau of Labor Statistics
Second, the table shows that the overall growth rate in Oregon employment was lower in 1997 than in 1996, a reflection that the state's strong economic growth had slowed somewhat. The data also show that the slowdown in the rate of retail trade employment growth from 3.1 percent in 1996 to 2.8 percent in 1997 was actually smaller than the decline in the growth rate for employment in all other industries from 4.2 percent to 3.5 percent.(6) If the minimum wage increase had adversely affected employment opportunities for low-wage workers, it is likely that the growth rate in retail trade employment would have fallen by a greater margin than the fall in the overall employment growth rate. Because the slowdown in job growth was more modest in retail trade than in other industries these data also suggest that the minimum wage increase was not a leading factor behind the modest slowdown in the state's economic growth rate in 1997.
The Oregon Employment Department did expect some reduced growth in retail trade employment as a result of the minimum wage increase. After factoring in the impact of the state's 1997 minimum wage increase, the department estimated that there would be 282,200 retail jobs in Oregon in 1997. That is one percent lower than the department's estimate of 285,100 retail trade jobs in the absence of the minimum wage increase.(7) Actual figures for 1997 which show that Oregon had 284,400 retail jobs, or slightly more than the number projected for the year support the Employment Department's forecast of modest job losses following the minimum wage increase.(8)
Research on the Minimum Wage's Effect on Employment
These indicators from Oregon are consistent with recent academic research showing that at the current level, moderate increases in the minimum wage would not have adverse employment effects. This includes studies of increases in the federal minimum wage in the early 1990s and in 1996, as well as increases at the state level.
Among the research on state minimum wages are studies of an increase in the California minimum wage in 1988 and in the New Jersey minimum wage in 1992. The New Jersey minimum wage was raised to $5.05 an hour in 1992. Adjusting for inflation, that equals $5.90 in 1998 dollars, or roughly the same as the 1998 Oregon minimum wage level. The study of the New Jersey minimum wage increase compared employment growth in fast-food restaurants in the state with employment growth in nearby eastern Pennsylvania, where the minimum wage was not raised. The study found that employment trends were as favorable in New Jersey as in Pennsylvania.(9) In response to questions about this study's methods, the authors prepared a 1998 re-analysis using new data from the Bureau of Labor Statistics. The new research confirms the initial findings.
Studies of recent increases in the federal minimum wage also have found no adverse effect on employment nationally. A May 1998 study by the Economic Policy Institute used four different tests to measure the employment effects of the 1996-97 increase in the federal minimum wage to $5.15 an hour. The tests typically showed small changes in employment for most groups in some cases positive and in other cases negative with virtually all of the changes being statistically insignificant. The one test that showed fairly substantial and statistically significant employment effects found that the minimum wage increase boosted the employment of low-wage workers. The authors concluded that the tests "fail to find any systematic, significant job loss associated with the 1996-97 increases."(10)
Implications for a Federal Minimum Wage
The Oregon findings have significant implications for the federal minimum wage debate. Just as in Oregon, parents who leave welfare for work throughout the nation tend to earn at or near the minimum wage. Nationwide, half of the single mothers who received welfare assistance for at least part of the year but also worked at least part of the year earned less than $5.15 an hour in 1996 the most recent year for which data are available while 63 percent earned less than $6.15 an hour. (The federal minimum wage, which stood at $4.25 an hour at the beginning of 1996 was raised to $4.75 an hour in October 1996 and to $5.15 an hour in September 1997. The leading proposal to raise the federal minimum wage would lift the wage floor to $6.15 an hour by 2000.)
In addition, the Oregon data are consistent with research showing that the recent increase in the federal minimum wage boosted the earnings of low-income families generally. A study by the Economic Policy Institute on the two-step increase in the minimum wage from $4.25 an hour in 1996 to $5.15 an hour in 1997 found that minimum-wage workers frequently reside in low-income working families and that low-income families benefited the most from the increase. The EPI study found that among working families with a working adult, those in the bottom fifth of the income distribution received five percent of total family income nationally, but they received 35 percent of the benefit from the minimum wage increase. The study also found that 58 percent of the earnings gains from the minimum wage increase went to the bottom two-fifths of families in terms of income. The EPI study also looked at the impact on wages for adults with less than a high school education and for teenagers; it found that there were substantial wage increases among these groups, particularly among minorities.(11)
Together, these findings suggest that the recent increases in the federal minimum wage are likely to have raised the incomes of a substantial share of the parents who left welfare for work. It further suggests that additional increases in the federal minimum wage would lift the earnings of these families even more, helping families make a successful transition to work and remain off welfare.
1. The ballot initiative called for a three-step increase in the state's minimum wage. The third step will be an increase to $6.50 an hour in January 1999.
2. The wage data reflect hourly wage information provided by welfare recipients to the welfare department. The large majority of recipients reporting wage information are participants in the state's welfare-to-work program. The remainder are recipients who find work outside of the state's JOBS program and report their wage information to caseworkers for benefit re-determination purposes. It is likely that the state data do not include all recipients who find work; for example, some recipients may choose to leave the welfare program when they find a job and thus would not need to report other wage information. Nevertheless, these data were collected in a consistent manner throughout the period covered in this report and thus are likely to accurately reflect changes over time. In addition, these data are used by the state as a performance measure for its welfare reform efforts.
3. Data from Oregon's Adult and Family Services division on the distribution of wages prior to the minimum wage increase are limited. They reflect only those welfare recipients who found full-time work, and the only figures on welfare recipients with earnings below the new minimum are for those earning less than $5.00 an hour. The available figures show that 17 percent of welfare recipients who found full-time work in the 1996 fiscal year earned less than $5.00 an hour. Since the vast majority of workers are covered by minimum wage laws, most of the workers earning less than five dollars an hour would have benefited from the state's minimum wage increase.
4. The national comparison group was selected to match as closely as possible the profile of Oregon welfare recipients who moved into work. The comparison group was limited to working women because most adult welfare recipients are women. The low-wage level is the 20th percentile wage level among all female workers, as calculated by the Economic Policy Institute. Some 80 percent of women earned more than this level and 20 percent earned less. Between 1993 and 1996, the 20th percentile wage level among women was slightly more than $6 an hour, or roughly equal to the average wage of Oregon welfare recipients who moved into jobs in that period.
5. The hourly wage for the typical low-wage female nationwide i.e., the 20th percentile wage among women equaled $6.13 an hour in 1993, $6.08 an hour in 1996, and $6.15 an hour in 1997, with all figures measured in 1997 dollars. It is worth noting that the increase between 1996 and 1997 may in part reflect the increase in the federal minimum wage from $4.75 an hour to $5.15 an hour in September 1997. In other words, the wage growth at this level may have been less than one percent if the federal minimum wage had not been increased.
6. When the change is measured in percentage terms, the 1997 growth rate in retail trade employment was 10 percent lower than the 1996 growth rate. The 1997 employment growth rate for all other industries was 15 percent lower than the 1996 growth rate.
7. Oregon Employment Department, unpublished data, March 1997. The simulation also estimated that when the minimum wage increase enacted in 1996 was fully implemented, i.e, when the state's minimum wage reaches $6.50 an hour in 1999, retail employment will be about three percent lower than it would have been if the minimum wage had not been increased. The simulation also estimated that the number of non-health service jobs will be about one percent lower in 1999 than it otherwise would have been as a result of the minimum wage increases. Overall, the department estimated that there would be 12,600 fewer jobs in the state in 1999 as a result of the minimum wage increase, a reduction of less than one percent.
8. As part of its standard measurement of employment, the Oregon Employment Department will revise all 1997 employment figures early in 1999. The revised figure for the number of retail trade jobs thus may differ slightly from the figure noted in this report.
9. David Card and Alan Krueger, "Minimum Wages and Employment: a Case Study of the Fast Food Industry in New Jersey and Pennsylvania," American Economics Review, September 1984, and "A Reanalysis of the Effect of the New Jersey Minimum Wage Increase on the Fast-Food Industry with Representative Payroll Data," Princeton University, January 1998.
10. Jared Bernstein and John Schmitt, Making Work Pay: The Impact of the 1996-97 Minimum Wage Increase, Economic Policy Institute, Washington, D.C., May 1998, page 1.
11. Bernstein and Schmitt, op cit. The study shows, for example, that 30 percent of working women aged 20 to 54 with less than a high school degree earned $5.15 an hour or less in the six months prior to the first step of the recent federal minimum wage increase in October 1996. During the six months following the second step of the increase, to $5.15 an hour, the share of teens with earnings below this level dropped to 13 percent. (According to EPI, some workers continued to earn below $5.15 despite the increase in the federal minimum wage because not all workers are covered by the minimum wage law, because some employers may not have been complying with the law, and because some workers may have inaccurately reported their earnings in the Census Bureau survey from which these data were drawn.)