Unionization And Working Poverty

Our guest author today is Ian Robinson, Lecturer in the Department of Sociology and in the Residential College’s interdisciplinary Social Theory and Practice program at the University of Michigan.

Poverty is (by definition) a function of inadequate income relative to family or household size. Low income has two possible proximate causes: insufficient hours of employment and/or insufficient hourly wages.  In 2001, there were four times more poor U.S. households in which someone had a job than there were in households in which no one did.  The same is still true today.  In other words, despite levels of unemployment far above post-World War Two norms, low wage jobs are by far the most important proximate cause of poverty in America today.

Perversely, despite this reality, the academic literature on U.S. poverty pays less attention to such jobs than it does to unemployment. A recent article, published in the journal American Sociological Review, both identifies and makes up for that shortcoming. In the process, its authors arrive at some striking conclusions. In particular, they find that unions are a major force for reducing poverty rates among households with at least one employed person.

Average poverty rates vary considerably among with the 50 states of this country. So do levels of unionization and other factors (e.g., the share of families headed by a single adult and the level of unemployment in the state) that figure prominently in most analyses of the deeper causes of working poverty. Thus, American federalism provides an enormous real-world experiment that we can use to assess the relative contributions of different factors affecting state poverty rates. The authors pool data on each of these factors, from the 50 states (and the District of Columbia), for a number of time slices between 1991 and 2010.  They then employ several different types of statistical analysis to arrive at the following conclusions:

• Higher levels of unionization in a state result in lower levels of poverty among most working families, both those with a union member, and the much larger number with no union members.
• Unions’ poverty-reducing effect does not flow solely from collective bargaining and the higher wages and benefits that it provides to union households; it also flows from a combination of:

a. the more worker-friendly government policies that stronger labor movements are able to secure (e.g., more generous earned income tax credit provisions, higher minimum wage, longer and better unemployment insurance benefits, etc.); and b. increased pressure on nonunion employers to pay higher wages in order to dissuade their employees from forming a union, or leaving to work at a higher paying union workplace.

• Union density levels (i.e., the share of the workforce belonging to unions) have a larger impact on working poverty levels than any of the other state level variables examined, including  GDP  per capita, economic growth rates, unemployment levels, and social policies [882-4].
• The poverty-reducing gains for working households -- union and nonunion -- that flow from higher levels of unionization are not achieved at the expense of higher levels of unemployment, contrary to the expectations of conservative economists [888].
• Three individual-level factors -- levels of education, the number of wage earners in the household, and the age of the primary income earner – have a bigger impact on levels of working poverty than union density.
• However, the first of these three factors, and perhaps the second as well, could be heavily influenced by public policies that vary with the strength of the labor movement.  For example, if we substantially increased government financial support for public higher education, tuition rates could be reduced and enrollments would increase.  The labor movement would likely push for an increase in public spending on higher education.  The authors recommend a systematic investigation of these possible relationships in future research.
• Perhaps the best way to explore these relationships is through international comparisons that allow us to assess what happens with education and other relevant policies in countries where the labor movement has much more political power than it does in any of the 50 states in this country.

Moreover, while working poverty can be reduced by increasing workers’ education levels, it is also true that the average low-wage worker has a substantially higher level of education today than she did 30 years ago. If the share of low-wage jobs in our economy grows, the share of low-wage workers will grow as well, even if average levels of education continue to increase. If you doubt this last proposition, consider the fact that 67% of all higher education faculty in this country were non-tenure-track by 2009, up from 22% forty years earlier.  Then ask nontenure track faculty -- most of whom have a Masters degree, and many of whom have a Ph.D. as well -- how much they earn!

The larger lesson of the lecturers in higher ed is this: if we want an economy that raises all boats and eliminates poverty, it will not be enough to raise education levels.  We will also need to change the policies that result in our economic system’s current tendency to generate more and more low wage jobs.

It will take a politically powerful labor movement to bring about those policy changes.  Attacking teachers and their unions in the hope that this will improve the quality of education, while assuming that better education is the key to escaping poverty, is thus a doubly misguided strategy. Of course, if destroying unions is the goal, and reducing poverty is only a fig leaf, the current discourse and strategy of the corporate education reformers makes excellent sense.

- Ian Robinson