Frank A. Scott

Bounding Parameter Estimates with Nonclassical Measurement Error

Abstract The bias introduced by errors in the measurement of independent variables has increasingly been a topic of interest among researchers estimating economic parameters. However, studies typically use the assumption of classical measurement error; that is, the variable of interest and its measurement error are uncorrelated, and the expected value of the mismeasured variable is equal to the expected value of the true measure. These assumptions often arise from convenience rather than conviction. When a variable is bounded, it is likely that the measurement error and the true value of the variable are negatively correlated. We consider the case of a noisily measured variable with a negative covariance between the measurement error and the true value of the variable. We show that, asymptotically, the parameter in a univariate regression is bounded between the ordinary least squares (OLS) estimator and an instrumental variables (IV) estimator. Further, we demonstrate that the OLS bound can be improved in the case where there are two noisy reports on the variable of interest. In the case of continuous variables, this lower-bound estimate is a consistent estimate of the parameter. In the case of binary or discrete noisily measured variables, we also identify point estimates using a method-of-moments framework. We then extend our bounding results to simple multivariate models with measurement error. We provide empirical applications of our analytical results using employer and employee reports on health insurance coverage and wage growth, and reports of identical twins on the level of schooling and wages. Using OLS, health insurance coverage is associated with a reduction in wage growth of 6.5–7.4%, whereas IV estimates suggest a 11.2–11.8% reduction associated with health insurance coverage. We are able to improve the lower bound estimate to 8.2% using our bounding strategy and obtain a point estimate of 8.8% using the method-of-moments framework. The estimates using the data for identical twins, though not correcting for problems such as endogenous determination of the level of schooling, do illustrate the potential usefulness of correcting for measurement error as a complement to other approaches. Using the multiple reports on the level of schooling and the our proposed estimators, we are able to tighten the spread between the upper- and lower-bound estimates of the returns to schooling from 7–10 percentage points to approximately 4 percentage points.

Effects of the Tax Treatment of Fringe Benefits on Labor Market Segmentation

Current federal tax law requires that all workers having the same experience with a firm must receive essentially the same package of fringe benefits in order for those benefits to qualify for preferential tax treatment. The authors argue that this “nondiscriminatory” provision of fringe benefits promotes labor market segmentation by inducing workers to sort themselves across the economy according to their demand for fringe benefits. Several empirical tests using 1968–78 data confirm the existence of sorting and suggest that labor market segmentation increased over time as workers responded to changes in incomes and marginal tax rates.

Is There Job Lock? Evidence from the Pre-HIPAA Era

We estimate discrete time hazard models of employment duration and standard logarithmic wage equations using the 1987 and 1990 panels of the Survey of Income and Program Participation (SIPP) to examine the phenomenon of job lock. We test for job lock using differences-in-differences approaches among those with and without employer-provided health insurance and family members with and without health problems. We find no statistically significant evidence of job lock on employment duration or wages using this approach. We do find some evidence of shorter employment spells for those with employer-provided health insurance and spouse-provided health insurance, and longer employment spells for those with employer-provided health insurance and large families. Others have interpreted these findings as evidence of job lock. However, the wage equation results using these measures are not consistent with job lock. Although anecdotal evidence makes it clear that some workers have been locked into less-than-optimal jobs because of the combination of health problems and employer-provided health insurance, our results do not suggest that this phenomenon is pervasive in the U.S. economy.

Health insurance coverage of the unemployed: COBRA and the potential effects of Kassebaum-Kennedy.

We use the April 1993 Current Population Survey to examine the health insurance coverage decisions of the unemployed and to simulate the potential effects of the new Kassebaum-Kennedy legislation. After controlling for demographic characteristics, COBRA eligibility raises the probability of health insurance coverage by 0.095, while eligibility for spouse employer insurance increases the likelihood of coverage by 0.318, and eligibility for both increases the likelihood of coverage by 0.341. In our simulations, we find that had Kassebaum-Kennedy been in effect in April 1993, 9.0 percent of the unemployed would be eligible to take up coverage, and the coverage rate of the unemployed would have been increased by 0.85 percent to 1.5 percent from 41.6 percent. Our estimates of the effect of Kassebaum-Kennedy on health insurance coverage are much lower than those reported by the Government Accounting Office prior to the passage of the legislation.

Concentration and Market Structure in Local Real Estate Markets

The competitiveness of the residential real estate brokerage industry has attracted much attention. Anecdotal evidence suggests some local markets are concentrated, yet no systematic market structure study has been conducted. We collected cross-sectional data on real estate brokers in 90 diverse markets across the United States and collected longitudinal data for Louisville, Kentucky. In medium and large markets, no evidence exists that market concentration might create problems for competition. Small markets, on average, have higher Herfindahl-Hirschman Indexes than medium and large markets. The longitudinal data reveal that many small brokers sell a house or two one year and none the next year.

Free agency, owner incentives, and the national football league players association

In an effort to raise salaries in professional football, the National Football League Players Association (NFLPA) has demanded that NFL owners set aside 55 percent of gross revenues to fund a union wage scale. An alternative means of raising salaries, as evidenced by the relatively higher earnings in major league baseball and professional basketball, is through a free agent system wherein players are free to sell themselves in the open market. This paper examines the NFLPA’s claim that free agency will not work in the NFL because owners lack the financial incentive to win that would induce them to bid on free agents.

Assessing the Intensity of Sports Rivalries Using Data From Secondary Market Transactions

Assessing the intensity of college football rivalries is an annual exercise of sports columnists. Fan polling, average ticket price, and even the bad behaviors of athletes have been used to quantify and rank rivalries. For economists, market prices of tickets have an appealing behavioral interpretation as they represent fans’ marginal willingness to substitute game attendance for other sources of utility. We have collected secondary market data on 278,117 individual ticket sales at different points in the season for 171 home games played by Southeastern Conference (SEC) member football teams over a 2-year period. Our rich data set allows us to control for the quality of the seats and the effects of on-field successes or failures during the season. Since we use data from 2 successive years, we are also able to normalize the level effect on prices of certain stadiums. We construct a willingness-to-pay measure of fan interest in various matchups and use this to compare the intensity of different rivalries for SEC schools.

Critical mass in the production of Ph.D.s: a multi-disciplinary study

Abstract Doctoral education in the US, while produced by not-for-profit institutions, is nonetheless carried out in a very competitive environment. We use survivor analysis to evaluate critical mass and minimum efficient scale in the production of Ph.D.s. We use data from the National Research Councils Survey of Earned Doctorates to study four academic disciplines: economics, history, physics, and psychology. We find that in economics, small programs are viable but intermediate sized programs are the norm. In history, a large number of very small programs exist, producing less than five Ph.D.s per year. Physics is characterized by somewhat larger scale production, which is not surprising given the nature of the production process in physics. Psychology seems to lend itself to large-scale production to a degree not evident in the other three disciplines. Size and quality ranking are positively correlated in economics, history, and physics, but no such relationship is apparent in psychology.

What Are SEC Football Tickets Worth? Evidence from Secondary Market Transactions

Tickets to sporting events are highly differentiatedoseat location, date and time of the game, and home-team and opponent qualities make each ticket unique. Preferences also differ nontrivially across fans, all of which make the suppliers pricing problem complex. We examine strategies employed by Southeastern Conference (SEC) universities in pricing their football tickets and evaluate their effectiveness in extracting surplus from fans. We use hedonic analysis of data collected from online secondary market transactions to construct a synthetic season ticket, which we compare to prices actually charged by university athletic departments. We also compare quality premiums charged by universities for better seats with market evaluations of those quality differences.

The Market for Real Estate Brokerage Services in Low- and High-Income Neighborhoods: A 6 City Study

We examine the market structure for real estate brokerage services across six large metropolitan areas, to see whether low-income neighborhoods or neighborhoods where house prices are low are as well served by real estate professionals as higher income or higher price neighborhoods. We collect more than 300,000 real estate listings and compute the Herfindahl-Hirschman Index (HHI) for each zip code neighborhood in each MSA. When we divide neighborhoods based on income, house value, and race, we find no evidence that access is worse in disadvantaged areas; that is, the market structure for brokerage services is at least as competitive in less advantaged neighborhoods. We also analyze market leaders in the six cities and find that some firms specialize in particular market segments, however.