Working Papers

Teacher Quality and Student Inequality

Richard K. Mansfield — Mon, 02 Jul 2012 11:41:14 PDT

This paper examines the extent to which the allocation of teachers within and across public high schools is contributing to inequality in student test score performance. Using ten years of administrative data from North Carolina public high schools, I estimate a flexible education production function in which student achievement reflects student inputs, teacher quality, school quality, and a school-specific scaling factor that allows the impact of teaching quality to vary across schools. The existence of nearly 3,000 teacher transfers, combined with a testable exogenous mobility assumption, allows separate identification of each teacher’s quality from both school quality and school sensitivity to teacher quality. I find that teaching quality is surprisingly equitably distributed both within and across high schools. Schools predominantly serving underprivileged students employ teachers who are only slightly below average, and most students receive a mix of their school’s good and bad teachers. Overall, I find that the allocation of teacher and school inputs at the high school level contributes only 4% to the achievement gap between the top and bottom deciles of an index of student background. Finally, I find that schools that disproportionately serve disadvantaged students tend to be more sensitive to teacher quality.

Thinking About Five Strategies for Making Diversity Work

Susan E. Woods — Tue, 01 May 2012 09:14:32 PDT

Greater diversity in the workplace introduces the possibility for increased misunderstanding and conflict at the same time as it holds out the promise of creativity and innovation. Workplace diversity change leaders have learned that making diversity work cannot be taken for granted. We cannot automatically assume that people will engage well with others across differences. Our history with bias, inequity and exclusion remains too much a part of how we understand one another. Making diversity work represents a mindset shift in the way people interact and engage. Organizations that are serious about creating inclusive work environments—where everyone feels welcomed, respected, and valued for who they are—recognize the importance of how people work together. Relationship building across differences needs to be developed and nurtured. Inclusive organizations focus on creating internal culture change to build people skills and promote shared expectations for mutual respect— an evolving endeavor. I’d like to add to the conversation by offering several observations for what this mindset shift might entail.

Thinking About LGBT Diversity in the Workplace

Susan E. Woods — Tue, 01 May 2012 09:11:21 PDT

[Excerpt] Today, according to a May 13, 2011 report by the PEW Research Center, “a majority of Americans, 58%, now say that homosexuality should be accepted, rather than discouraged by society.” There are many reasons for this more inclusive shift in attitudes. Inclusion of sexual orientation, and increasingly gender identity, in workplace inclusion initiatives and diversity awareness dialogue is likely one. The workplace, driven by the pragmatic need for improved productivity, talent recruitment, and retention of a motivated workforce, has become a powerful environment for social change and learning.

The Empirical Case for Streamlining the NLRB Certification Process: The Role of Date of Unfair Labor Practice Occurrence

Kate Bronfenbrenner et al. — Wed, 11 Apr 2012 13:43:58 PDT

[Excerpt] One of the long held performance objectives of the National Labor Relations Board (NLRB) has been to reduce the time period between the filing of the petition and union certification elections. This year the NLRB's 2010 Performance Accountability Report claimed that 86.3 percent of all NLRB elections were held within 100 days of the petition being filed and 95.1 percent of all initial elections were held within 56 days of the petitions being filed. Our analysis of Bureau of National Affairs (BNA) data from 1999-2009 found that in the last two years there has been a slight increase in the number of representation elections being held between 21-30 days after the petition. But throughout the decade there have been virtually no election dates in the first 20 days after the petition is filed (see Figure 1). Thus, while the NLRB has made some progress in meeting their performance objectives, as former NLRB General Counsel Fred Feinstein explains, "the problem has been that a party in any election case has the ability to undermine the expression of employee free choice by manipulating the Board procedures to create delay”.

These data on the timing of petition and election dates matter because the number of days between the petition and the election may make a great deal of difference as to whether or not a group of workers get union representation or a first contract. The time between the petition and the election also may make a difference as to whether an individual worker gets fired or gets his or her wages or benefits cut in retaliation for union activity, or ends up leaving a job, with or without a settlement, because of the coercion, threats, and/or harassment suffered throughout the employer's campaign. Over the last three decades there has been a wealth of scholarly research that has documented the gauntlet of threats, fear, retaliation, misinformation, and harassment workers have to endure in order to exercise their right to union representation and collective bargaining.

However, none of the recent NLRB certification election research has examined the relationship between employer opposition and the timing of the election. This is because for the few researchers who have done the hard work of connecting election to unfair labor practice (ULP) data, the only timing variable they had to use for employer behavior was the date ULPs were filed. But since ULP charges can be filed as much as six months after the allegation occurred, "date filed" grossly underestimates when employer opposition begins. It is for this reason we were asked to follow up on our 2009 study on employer opposition and the breakdown of the unfair labor practice process (see No Hold Barred: The Intensification of Employer Opposition to Organizing) to examine the relationship between the petition date, election date, and when the most serious employer opposition occurs during representation campaigns. Upon completion of this analysis we would be able to determine whether our data provide empirical support for significantly reducing the number of days between petition and election in NLRB certification elections.

Union Tactics Matter: The Impact of Union Tactics on Certification Elections, First Contracts and Membership Rates

Kate Bronfenbrenner et al. — Mon, 05 Mar 2012 08:40:21 PST

[Excerpt] This study examines the impact of union tactics on certification election win rates, first contracts and post-contract membership rates in the public sector. Based on an in-depth survey of union organizers in a national sample of public sector certification election campaigns the findings suggest that a grassroots, rank-and-file intensive strategy, building a union and acting like a union from the very beginning of the campaign are critical components of union organizing success. By comparing these findings with Bronfenbrenner's earlier study of private sector union campaigns (1993) we found that not only are these strategies important to the union's ability to win elections and first contracts, but they are equally important to the union's ability to sign up new members, and build a lasting, viable organization, after the election is won and the first contract is signed.

Labor Market Analysis for Developing Countries

Gary S. Fields — Thu, 11 Aug 2011 12:44:01 PDT

This paper is about analyzing labor markets in developing countries, searching for both improved understanding and greater policy relevance. Following a five-part policy evaluation framework, the highlights of labor markets in developing countries are presented. Theoretical models with multiple sectors and segments and empirical analysis using different kinds of data are then reviewed. A brief concluding section addresses some priority research needs.

Poverty Effects of the Minimum Wage: The Role of Household Employment Composition

Gary S. Fields et al. — Thu, 11 Aug 2011 12:41:38 PDT

A change in a country’s minimum wage will in general affect the number of workers in covered sector employment, uncovered sector employment, and unemployment. The impact of these labor market adjustments on absolute poverty will depend on how the pattern of employment composition changes within households and on how income is shared within households. An earlier paper (Fields and Kanbur, 2007) focused on the income-sharing dimension of the problem. The present paper focuses on household employment composition. For a particular structure of the labor market— one with good jobs, bad jobs, unemployment, and adult and youth workers— and with a particular model of how the sectoral patterns of employment are translated into household employment composition, we analyze the impact of minimum wages on a class of absolute poverty measures. The precise characterizations demonstrate the need for a nuanced appreciation of the impacts of a minimum wage increase, since they depend intricately on the values of key parameters (the poverty line, poverty aversion, labor demand elasticity, and the starting level of the minimum wage). Moreover, the relationship between poverty and the minimum wage is in general non-monotonic, so that local effects can be quite different from the effects of large changes in the minimum wage.

But That’s Not What Economic Mobility Is!

Gary S. Fields — Thu, 11 Aug 2011 12:37:10 PDT

[Excerpt] How many times have you attended a talk or read a paper on economic mobility and thought, “What you are talking about is not what I am interested in”? Not only do different people have different ideas about what economic mobility is, but they have different clear ideas about what economic mobility is. The purpose of this paper is to present the essential features of the different economic mobility concepts that are found in the literature.

What We Know (and Want to Know) About Earnings Mobility in Developing Countries

Gary S. Fields — Thu, 11 Aug 2011 12:34:38 PDT

[Excerpt] Some developing countries have experienced rapid economic growth, some slow economic growth, some no growth at all, and some economic decline. The traditional way of gauging the distributional consequences of economic growth, if in fact there was economic growth, is to use data from comparable cross sections to calculate various measures of (relative) inequality and (absolute) poverty. The very large literature on inequality and poverty will not be reviewed here.

A newer approach in the development literature is to study the distributional consequences of economic growth (or non-growth) by using data for the same recipient units for two or more points in time to analyze changes in total income (“income mobility”) and in income from paid employment and self-employment (“earnings mobility”). Such data, called panel data or longitudinal data, may involve baseline interviews and one or more subsequent reinterviews or alternatively a single interview with retrospective questions about previous income or earnings. Examples of panels with reinterviews are South Africa’s KwaZulu-Natal Income Dynamics Study, the Indonesia Family Life Study, and Chile’s CASEN panel. A prominent panel based on retrospective data is the China Household Income Project. The literature reviewed in this paper draws on both kinds of panel data.

Reflections on My Immersion in India

Gary S. Fields — Thu, 11 Aug 2011 12:32:05 PDT

[Excerpt] Kalavati’s family derives all of its money from labor earnings. Her husband long ago worked for the Cannon textile mills when they were still making bath towels in India. Sixteen years earlier, the mill shut down, moving to a place where labor was even cheaper, and her husband lost his job. For fifteen years, he did not work. Then finally, he got a job where he works at night. He does not tell Kalavati where he works or how much he earns, nor does he contribute his earnings to the day-to-day expenses. (He does contribute to interest payments to the local money lender, though.)

Poverty and Low Earnings in the Developing World

Gary S. Fields — Thu, 11 Aug 2011 12:30:06 PDT

More than three billion people are poor by international standards, and essentially all are to be found in the low- and middle-income countries of Asia, Africa, and Latin America. The issues for understanding poverty in the developing world - among them, self-employment and household enterprises, agricultural work, casual employment, and informal work – differ from those in the developed world. Different policy issues predominate: stimulating economic growth, harnessing the energies of the private sector, increasing paid employment, and raising the returns to self-employment. This chapter details how the poorer half of the world’s people work and gives an overview of lessons from around the world on what has helped improve their earning opportunities. The chapter concludes with suggestions for future research.

Community College Transfer Students’ Probabilities of Baccalaureate Receipt as a Function of their Prevalence in Four-Year Colleges and Departments

Andrew W. Nutting — Thu, 31 Mar 2011 08:15:19 PDT

I determine whether community college transfer students have higher baccalaureate rates when they enroll in four-year colleges and departments that have larger shares of transfer students. Transfers attending non-technical campuses with larger shares of transfers have higher eight-year baccalaureate rates, but within-campus increases in share transfers do not increase transfer graduation rates. Transfers in departments with large shares of transfer students have significantly lower graduation rates, but natives in such departments do not. Within-department increases in transfer student presence are positively correlated with transfer eight-year graduation rates and negatively correlated with native eight-year graduation rates, indicating an opportunity for efficiency gains if influxes of transfers are separated from natives.

The Graduate Education Initiative: Description and Preliminary Findings

Ronald G. Ehrenberg et al. — Thu, 31 Mar 2011 08:15:16 PDT

[Excerpt] In1991 the Andrew W. Mellon Foundation launched the Graduate Education Initiative (hereafter GEI) to improve the structure and organization of PhD programs in the humanities and social sciences. Such changes were seen as necessary to combat high rates of student attrition and long times-to-degree in these programs. While attrition and time-to-degree were deemed to be important in and of themselves, and of great significance to degree seekers, they were also seen more broadly as indicators of the effectiveness of graduate programs. Several characteristics of doctoral programs were earmarked as contributing to high attrition and long degree time, including: unclear expectations, a proliferation of courses, elaborate and sometimes conflicting requirements, intermittent supervision, epistemological disagreements on fundamentals and not least, inadequate funding. Projections that faculty shortages would occur in the late 1990s in the humanities made the goals of reducing student attrition and time-to-degree particularly timely if an adequate number of PhDs were to be available.

This was far from the first such effort to reduce times-to degree-and rates of attrition. Earlier programs, which provided grants in aid to individual students or to graduate schools to distribute as they saw fit, had failed conspicuously. Based on data which showed that there were marked differences among departments and on a great deal of experience on the ground, the architects of the GEI determined that to improve graduate education would require departments to make changes in their PhD programs. As such, the Foundation shifted much of its support for doctoral education, which had previously gone directly to students, to block grants that would be awarded to departments within major universities.

Does Public Funding for Higher Education Matter?

Liang Zhang — Thu, 31 Mar 2011 08:15:13 PDT

This study uses panel data to examine the direct link between state funding and graduation rates at four-year public institutions. When other factors are held constant, a $1,000 increase in state appropriations per FTE student at four-year public institutions is associated with about a one percentage point increase in graduation rates. This positive link appears to hold for all research/doctoral, masters, and baccalaureate institutions. In addition, there is evidence that modest increases (or a decrease) in state funding are associated with rapid increases in tuition rates charged at four-year public institutions, which likely result in an additional negative impact on graduation rates. Simply put, there is no such a thing as free lunch when it comes to graduation rates at public higher education institutions.

Money Changes Everything: Funding Shocks and Optimal Admissions and Financial Aid Policies in Higher Education

Matthew G. Nagler — Thu, 31 Mar 2011 08:15:11 PDT

The paper examines the effect of a shock to university funding on tuition net of financial aid, admissions selectivity, and enrollment levels chosen by an optimizing university. Whereas a positive shock, such as a major donation, results in lower net tuition and greater selectivity with respect to all students, its effect on enrollment may not be uniform. Student categories given little weight in the university’s utility function may be treated as “inferior goods,” that is, their enrollment may be decreased in the face of a positive shock, while other student categories see enrollments increased. Such students are charged net tuition well above their marginal cost of enrollment and play primarily a revenue- rather than prestige-generating role for the university. Inferences are drawn from the analytical framework concerning the effect on tuition levels of an increase in federal direct-to-student aid, permitting a new perspective on evidence relating to the Bennett hypothesis.

Does a Spouse Slow You Down? Marriage and Graduate Student Outcomes

Joseph Price — Thu, 31 Mar 2011 08:15:06 PDT

Using data on 11,000 graduate students from 100 departments over a 20 year period, I test whether graduate student outcomes (graduation rates, time to degree, publication success, and initial job placement) differ based on a student’s gender and marital status. I find that married men have better outcomes across every measure than single men. Married women do no worse than single women on any measure and actually have more publishing success and complete their degree in less time. The outcomes of cohabiting students generally fall between those of single and married students.

Gender Differences in the Response to Competition

Joseph Price — Thu, 31 Mar 2011 08:15:03 PDT

I use the introduction of a competitive fellowship program for graduate students to test whether men and women respond differently to competition and whether this response depends on the gender mix of the group. Men experienced a 10% increase in performance in response to the program, with the largest gains for men in departments with the most female students. Women did not increase performance, on average, but the response of women did differ greatly depending on the gender mix of their peers, with a more positive response when a larger fraction of the group was female.

Citizenship, Gender, and Racial Differences in the Publishing Success Of Graduate Students and Young Academics

Joseph Price et al. — Thu, 31 Mar 2011 08:14:59 PDT

Although extensive research exists on the publishing success of academics, few studies have examined factors influencing the publishing success of graduate students and young academics. Data from a survey of 12,000 graduate students in the Humanities and related social sciences was used to examine citizenship, gender and racial/ethnic differences in publishing success during graduate school and the first three years after graduation. The results of this analysis indicate that international students have the highest publication rates during graduate school as well as in the first three years following receipt of degree. Results also indicate that female graduate students are less likely than male graduate students to publish, a gap that remains in the years following graduate school. Finally, results indicate that U.S. citizen minority students exhibit lower levels of publishing success compared with non-minority students during graduate school, but that this gap that disappears within the first few years after graduate school.

Faculty Employment and R&D Expenditures at Research Universities

Liang Zhang et al. — Thu, 31 Mar 2011 08:14:56 PDT

This study uses panel data to examine the relationship between faculty employment and external R&D expenditures at research and doctoral institutions over a 15-year period of time. Not surprisingly, full-time faculty that are tenured or on tenure-tracks is the main category of faculty that generates external R&D funding. On the other hand, our results suggest that an increasing usage of part-time faculty, holding constant the institution’s full-time faculty size boosts an institution’s external R&D expenditures, probably through reducing teaching responsibilities for the full-time faculty. Increases in graduate student enrollments are associated with increases in external R&D expenditures. Finally, an institution’s external R&D expenditures are significantly influenced by both the amount of its own institutionally financed research expenditures and the level of federal funding for research.

Transfer College Quality and Student Performance

Angela K. Dills et al. — Thu, 31 Mar 2011 08:14:53 PDT

With escalating tuitions, the return to college quality remains an important consideration for students deciding whether to spend more money for a higher quality college education. This paper examines how students that transfer credit for an introductory-level course perform in a subsequent intermediate-level course. Using rich administrative data we estimate how college quality affects student performance, holding constant many observable student characteristics.

Students taking introductory courses at higher quality institutions earn higher grades in their intermediate courses than students attending lower quality institutions. This difference is small, but statistically significant. A back of the envelope calculation suggests that, for the average student, the benefit from attending a higher quality institution is more than worth the higher tuition.