The Flip Side of Turnover: Employment Transitions and Occupational Attachment Among Low-Wage Care Workers in the United States

Abstract

Scholars have explored the ways that conventional economic theory does not fully explain the distribution and characteristics of caring labor – the work, unpaid and paid, of caring for those who are young, elderly, or disabled. This paper explores a critical dimension of paid care – high turnover rates in the lowest-wage segment of the sector (including childcare, nursing homes, home health). Using longitudinal data from the 2008 Survey of Income and Program Participation (SIPP) in the US, it examines the “flip side” of that turnover by comparing occupational mobility among low-wage care workers to that of other low-wage service workers. The findings indicate that patterns of occupational transition among care workers are distinct in important ways. Understanding occupational attachment among paid care workers is critical to developing theoretical models about care and to creating care-specific policies to address employee turnover and its negative impact on care quality.

HIGHLIGHTS

• High turnover in low-wage jobs in care-related fields has a negative impact on the quality of care.

• Low-wage care workers have longer job tenures and are more likely to stay in their field than other low-wage workers.

• Low-wage care workers experience more upward mobility than food service and cleaning workers, but less than office and sales workers.

• Low-wage care workers have high rates of transition to a period of not working.

• There is evidence of higher levels of occupational attachment among low-wage care workers than among other low-wage service workers.

KEYWORDS:

• Care
• low-wage employment
• occupational mobility

JEL Codes:

• I1
• J6
• B54

Notes

1 We have deliberately chosen to discuss “low-SES” workers rather than “low-skill” workers because of the evidence of the social construction of skill and its particular relevance to female-dominated occupations (Steinberg 1990). We use the terms “low-wage workers” and “low-SES workers” interchangeably for ease of readability – in all cases we are referring to workers identified by this more expansive set of criteria besides wages.

2 Robert M. Hauser and John Robert Warren (1996) base occupational prestige on the 1989 prestige ratings calculated by Nakao and Treas (1994) using the 1989 General Social Survey. Occupations are rated using a 9-point scale and the prestige measure reflects the percent surveyed who rate the occupation on the scale.

3 For ease of reading we will refer to them in shorthand for the remainder of the paper: Food Preparation and Serving Related Occupations as “Food service”; Building and Grounds Cleaning and Maintenance as “Cleaning”; Sales as “Sales”; and Office and Administrative Support as “Office.”

4 We drop 221 spells in which a worker at any point in the spell has an imputed wage of over $60/hour because these skew our wage statistics (primarily for care workers). Dropping them does not change the pattern of results in subsequent analyses and tables.

5 While this appears to contradict earlier studies that have found a wage penalty for low-wage care workers, it is important to remember that in most wage penalty studies, the unit of analysis is the worker, and an individual worker is being compared to other workers who are similar in level of education and other characteristics. Our comparison is at the level of the occupation and here does not control for worker characteristics.

6 See Ham, Li, and Shore-Sheppard (2009) for a discussion of the advantages of this model.

7 We have also estimated these models on the full sample of spells (including truncated ones) and the results are very similar.

8 Note that this model is a formalization of the descriptive analysis above and simply models where workers go, conditional upon leaving a job. It is not a duration model with competing risks.

9 More technically, there are two omitted factors in the ML model; the occupation is care and the transition is staying in the same occupation. So all coefficients for the other occupations are relative to what care workers do. We see from descriptives that care workers are more likely to go back to the same occupation. And, just as importantly, office workers are unlikely to make that transition. So because office workers are less likely than care workers to stay in the occupation relative to other options, the estimated effect is negative effect, and the magnitude is large because of how low the office worker “stay” rate is (see Table 6).

10 As a specification check, we have run models with only healthcare jobs relative to the other occupation groupings, dropping childcare workers and social-service workers from our care-workers category. However, the model results with the healthcare sample are extremely similar to those in our baseline models.

11 One such factor could be health insurance. As we show in Table 1, care workers have the highest rate of coverage across the four groups. However, in results not reported (available upon request), we find that controlling for health insurance coverage in the multinomial logit model does not change the pattern of results.

Additional information

Notes on contributors

Mignon Duffy

Mignon Duffy is Associate Professor of Sociology at the University of Massachusetts Lowell. Her scholarship focuses on the paid care labor force and in particular its intersections with race, class, and gender inequalities.

Reagan Baughman

Reagan Baughman is Associate Professor of Economics in the Paul College of Business and Economics at the University of New Hampshire. Her research interests span a range of topics in health and labor economics, including health insurance coverage patterns, economic determinants of child health, evaluation of income support programs, and labor supply in the long-term care sector.

Kristin Smith

Kristin Smith is Visiting Research Associate Professor of Sociology at Dartmouth College. Her research focuses on gender inequality, employment and earnings, and work and family policy. Her expertise lies in examining gender differences in how work and family life interconnect and developing workplace policy recommendations.