Search

This study investigated whether overhaul costs differ in public and private shipyards, using data on overhauls of the Sturgeon class of nuclear attack submarines between 1971 and 1979. We began by estimating a cost function and using the regression coefficients to estimate the production cost of a hypothetical overhaul for each shipyard in each year. For the most part, it was found that estimated naval shipyard costs did not differ significantly from estimated private yard costs. There were, however, several instances in which naval yard costs were significantly higher than private yard costs. Among the reasons for the differences in estimated production costs in our sample were higher wage rates and a lower cost of capital in the naval shipyards. Experience levels also differed among shipyards, but were not consistently higher or lower for the naval shipyards. There may be a difference in overhaul quality, but such a difference did not show up in the measures of material condition that we examined. One difference that did emerge strongly concerned the length of time per overhaul, which was lower in the naval shipyards.

Data have recently become available that can be used to shed light on the determinants of rare but catastrophic diseases such as leukemia. These data are accounting-type records from the unemployment insurance systems in particular states. Because these data cover most workers (whether unemployed or not), they provide the huge samples needed to analyze these diseases. Their combination with health data and their role relative to other data sets are described in this paper. (Author)

The repair of naval vessels is one type of production which is performed in both the public and private sector. Input prices are not the same in naval and private shipyards; in particular, wages are higher and the cost of capital is lower in naval shipyards. In this paper, an attempt was made to determine how these differences in factor prices affect the relative cost of overhauls, and some possible reasons for differences in overhaul costs. First, a cost function was estimated, using data on overhauls of the Sturgeon class of nuclear attack submarines. The regression coefficients were then used to predict costs in public and private yards holding the relevant variables constant. The results support the hypothesis that, for the type of work studied, production costs were higher in naval shipyards than in private shipyards. Among the likely reasons for the difference in production costs are (a) higher wages; and (b) lower efficiency in naval shipyards. Further work is needed to determine whether there is also a difference in overhaul quality. The time spent in overhaul is important because of its implications for readiness. We found that overhauls in private shipyards took significantly longer than those in naval yards. When determining where to assign overhauls, the Navy must weigh this difference in time costs against the difference in production costs.

Technical change, for all the good it does for society, is not an unmixed blessing. Though it leads to the development of useful new products and new production processes, it may impose hardships on those who use old, and no longer efficient, methods or produce products that are no longer wanted. The net effect of technical change on workers is hard to predict. They can gain if their industries gain in competition with producers of similar products. As consumers, they also gain from increases in productivity; they are able to buy things at lower prices. But if workers cannot adapt to new production methods and lose their jobs as a result, they can end up as net losers

In this paper, we report on an investigation of the political value of federal spending in congressional elections. The basic hypothesis can be stated simply: An incumbent congressman's reelection prospects are positively related to changes in federal spending in his district. To test this hypothesis, we focus on two relationships: the dependence of the incumbent's vote on changes in local federal spending, expressed in percentage terms, and the dependence of campaign expenditures (for the incumbent and the challenger) or changes in local federal spending, expressed in percentage terms. These relations are examined on the basis of data from individual districts for each of three elections: 1980, 1978, and 1976.

This study estimates the earnings losses of workers who lose their jobs in a plant closing. A unique data set was used: Social Security earnings records of over 9,000 workers employed in plants that actually closed. Separate estimates are made for workers by age and sex and the effects on losses of economic and demographic variables are also estimated. Alternative methodologies are discussed and used to estimate losses of workers who never work after the plant closing. (Author)

In this paper, I assert that the country as whole, and the D.C. area in particular, would be better off if the federal government were to sell its airports and leave their operation to individual discretion, with no interference whatsoever. I'm going to support that assertion in two ways: first, I'll go over the conditions under which it would be appropriate for the federal government to administer airports and show that those conditions do not exist in the District of Columbia. Then, I'll compare the existing arrangement with what would happen of the airports were privately operated. (Author)

This paper estimates the value employees place on stable employment. Here the term 'stable employment' means a relatively low probability of temporary and/or permanent layoffs. This value is estimated by regressing individual wage rates on exogenous variables and proxy variables for unstable employment. The sign and size of the coefficients on these proxy variables in the wage equation measures the value of stable employment in terms of the hourly wage rate. The wage equation is estimated using the Michigan and Parnes survey data. The results indicate that the wage elasticity with respect to instability is .3. This means that if one industry is 50 percent more stable than another, then other things equal, the more stable industry would have a 15 percent lower wage rate. (Author)

The government can not base the amount it pays its employees on the market value of the services it provides. Instead, it tries to set pay for each job at a level equal to pay for comparable work in private employment. Although a survey of pay for private sector workers is taken each year, there are indications that the survey is faulty and that government pay levels are too high. One indication is that quit rates are considerably lower in the government than in the private sector. Quit rates are a function of pay levels. Other things equal, comparable quit rates indicate comparable pay. The evidence presented in this paper shows that one reason for the lower quit rates in the government is that the government is such a large employer. Large firms have lower quit rates, probably because transfer and promotion are easy within a large firm. Even when firm size is taken into account, however, the quit rate in the government remains lower than the quit rate for equivalent private sector workers, suggesting that compensation (pay plus other benefits) for government is above private compensation. The differential--shown in this paper to be 15 percent--is in line with findings in other studies that make direct pay comparisons, but these results must be considered tentative until other factors that influence the quit rate have been included in the analysis.

A number of studies have concluded that the demand for labor depends on both current and expected future output. Though many of these studies rely on the assumption of rational expectations, none has recognized and made use of a unique characteristic of the theory of rational expectations: that in the process of generating measures of expectations information is automatically created on the extent to which different variables affect expectations. Hence, the differing response to alternative causes of output change can be estimated. In this paper, we focus on imports as a distinct cause of output change. We test the hypothesis that employment reacts more rapidly to output changes when they are due to changes in imports than when they are due to the business cycle or other influences. It has been standard practice to use input-output studies to predict the effect of imports on the domestic demand in specific industries. The input-output model assumes that a change in output will cause proportional and immediate effects on industry employment, no matter what caused the change in output.

Technical change, making possible increased output from existing resources, enhances the potential standard of living for society. yet in the particular industries where technical change is occurring, there are conflicting effects: (1) the price of output will fall so that more is demanded and (2) the demand for some or all inputs for given output will be reduced. With respect to labor, the first effect increases demand, the second diminishes it. This paper presents estimates of the net effect of technical change on labor demand in four industries: steel, autos, aluminum, and iron ore. Using an economic model of these industries (which takes account of the linkages among them), we estimate what would have been the effect on employment if no technical change occurred in each since 1959. We find that if technical change had been suppressed, output and employment in two of the industries (aluminum and iron ore) would have been drastically reduced, largely because of an inability to compete with imports. The other two (steel and autos) would have suffered large drops in output, but their employment would be changed by less than 4 percent.

In setting the speed limit, the government should consider only externalities, e.g., the tendency of one driver's speed to endanger other drivers. In this paper, we discuss the methodology for estimating the optimal speed limit, based on externalities. We carry out a numerical illustration to demonstrate how the methodology leads to (1) a numerical estimate of the optimum speed limit, (2) an estimate of the dollar loss from a suboptimum speed limit, (3) an estimate of cost per life saved, and (4) the suggestion that federal traffic data collection efforts and traffic studies be redirected toward discovering two crucial parameters: the speed drivers would go if left alone and the ratio of the external to internal marginal cost of highway speed. (Author)

The effects of imports on industry employment are often determined through the use of input-output studies. Input-output assumes that imports cause proportional and immediate effects on industry employment. Increases in imports will therefore be expected to cause large, sudden decreases in employment. The problem arises, however, that actual events are often poorly predicted by the input-output model. To better predict the effects of imports on employment, a model of the demand for labor was developed that allowed for gradual adjustment in employment to perceived changes in output, where these changes arise either from cyclical factors or an increase in competing imports. What is expected to be produced in the future was felt to be an important determinant of current employment needs and therefore was explicitly included in the labor demand model. According to our findings, expectations of future output are important determinants of industry employment demand in the majority of industries studied. Perhaps, more surprisingly, imports induce a slower adjustment in employment than does an equivalent change in GNP, the measure used to represent cyclical factors. Our results suggest input-output studies overestimate the effects of competing imports on employment in the industry. (Author)

PRI conducted a study for the National Commission on Unemployment compensation to help UI administrators evaluate tax systems and predict how changes in tax systems would change fund balances. Several models of UI tax systems were developed, and their predictive power was tested. The best models were used to predict fund balances under a variety of hypothetical tax systems. The adequacy of fund balances was also evaluated. Federal administrators now consider a fund adequate if the balance is a specified multiple of annual wages 'at the beginning of a period of relatively high unemployment.' Aside from the difficulty of determining when 'relatively high unemployment' begins, this measure treats solvency as the only criterion for evaluating fund balances. States can keep their funds solvent by maintaining arbitrarily high balances or by raising taxes immediately when benefit outflows start to rise, but neither method is desirable. Very high balances represent idle funds that could have been used productively by the employers who pay unemployment insurance taxes and raising taxes when benefit outflows begin to rise means high taxes in a recession. To reflect the actual--often competing--goals of UI fund managers, the study team evaluated UI tax systems by three criteria: the mean balance, the probability of insolvency, and the timing of the tax over the business cycle. A good tax system will produce a fund with a low mean balance and a low probability of insolvency, and will raise taxes in a boom rather than in a recession.

Workers who are displaced when a plant closes, or when there is a permanent layoff resulting from reduced demand, usually suffer losses in earnings. These losses are due to unemployment and to wage reductions which reflect more permanent impairment of earnings capacity. It is important to know the long terms effects of job displacement. Many prospective changes in government policy can result in reductions in demand for specific products and lead to some job displacement. This paper reviews seventeen recent studies of the effects of job loss on earnings. They range from case studies of specific plant closures to more broadly based studies of the effects of job loss, in which control groups are used to estimate losses in earnings over a number of years after layoff. Methods of analysis range from simple tabulations to estimation of income determination models based on human capital theory. The losses documented in these studies can be taken to illustrate the magnitude of losses that would result from displacements due to increased import penetration. Some of the studies deal with job losses specifically due to increased import competition, but all are concerned with job losses following declines in demand for domestic production. Whatever the cause of a fall in demand for the products of an industry, increased import penetration, changes in tastes, etc., the skills of workers that are specifically adapted to either the industry or to particular firms within that industry become less valuable.

This study addresses the question of whether financially stressed airlines are likely to cut back activities contributing to airline safety, to reduce maintenance expenditures, or degrade service levels. Theoretical analysis shows that financially unsuccessful airlines have some incentives to cut back in these areas. The forces influencing such decisions are so numerous and complex, however, that we cannot predict whether such cutbacks would or would not occur in any particular case. We do show, however, that variations in the profitability, liquidity, and debt-equity leverage of the eleven trunk airlines over the period 1965 to 1977 did not affect their accidents, maintenance expenditures, and passenger complaints. A summary of this study and our findings are presented in this report. (Author)

In this paper, a method is proposed for separating the error term of the stochastic frontier model into its two components for each observation. This method enables one to estimate the level of technical inefficiency for each observation in the sample, and largely removes what had been viewed as a considerable disadvantage of the stochastic frontier model relative to other models (so-called deterministic frontiers) for which technical inefficiency was readily measured for each observation.

This paper contains the results of a preliminary investigation into the role that taxes may have played in the inflationary process of the past two decades. Such an analysis seems timely, because the proportion of GNP devoted to taxes has been rising fairly steadily since 1950. Moreover, not much attention has been paid in the literature to taxes in the possible role as causes of inflationary pressures. In recent macroeconomic textbooks, taxes have been analyzed in a predominantly Keynesian framework, in which taxes are determinants of aggregate demand. Accordingly, increases in taxes lead to reductions in aggregate demand and, hence, their impact tends to be deflationary. But aggregate supply remains completely unaffected by changes in taxes.

In 1970, about 500 occupations were licensed by one or more states in the U.S. and over 2000 were regulated at the local level. The fact that licensing is so widespread suggests that it must provide some benefits to the public, perhaps by eliminating from the market some low quality services that most consumers feel they would only have bought by mistake. It may also result in higher prices, however, without improving the quality of services available.

Between 1978 and 1985, municipal governments and private industry will spend about $111 billion dollars on construction of new facilities to reduce air and water pollution as mandated by the EPA. This report presents estimates of how these projected expenditures will affect the total level of construction and how employment, wages, interest rates, and other variables will respond to the change in construction output. Effects on employment in selected sub-classes of the construction industry were also estimated. The response of the construction industry to increased spending for pollution control was estimated by use of an econometric model of the industry. The model, developed for this study, was simulated to predict the effects of projected spending through 1985. The results suggest that construction output will rise by less than the EPA mandated expenditure. The difference is due, in part, to displacement of spending that local governments would otherwise have done and, in part, to absorption of the increase by rising wages and prices in construction. There is also some displacement of industrial expenditures for pollution abatement because of the increased cost of building a new plant. The results also suggest that employment in the construction industry is increased. Employment of young, black, inexperienced workers increases more than would be expected based on their representation in the construction labor force. (Author)

The research reported here is an extension of a previous enquiry by Brechling and Jehn into the relationships between labor turnover and the unemployment insurance tax. First, the data base was extended through 1977, and the original models were re-estimated. The original and enlarged data sets yielded similar conclusions. Second, the influence of the taxable wage base of the social security tax upon labor turnover was examined. This taxable wage base seems to have a different impact from that of the taxable wage base of the unemployment insurance tax. Third, the relationships between the seasonal pattern of labor turnover and the unemployment insurance tax were examined. The results tended to be weak. Viewed as a whole, the results of the new investigation are not as strong as the original ones, but they do support the conclusion that there are significant relationships between labor turnover and the parameters of the unemployment insurance tax. (Author), Continuation of Contract J-9-M-6-0103.

The Arizona Employment and Unemployment Data Set (AZEUD) is made up of information drawn from administrative records of government agencies in two separate samples: A 20% sample of workers who applied for unemployment insurance (UI) in Arizona 1963-71; and A 1% sample of the Arizona labor force 1957-72. The 20% sample is drawn from the Continuous Wage and Benefit History (CWBH) data, a data set maintained by the Arizona Bureau of Employment Security (AZBES). This data set contains detailed longitudinal histories of both the employment and unemployment experience of workers who claimed UI benefits in Arizona, 1963-71. In total, about 40,000 workers are in the 20% sample. The 1% sample is derived from the Social Security Administration's Longitudinal Employee-Employer Data (LEED) file. The data file contains detailed longitudinal histories of the employment experience of workers covered by Social Security employed in Arizona 1957-72. There is no explicit indicator of whether or not workers in this file claimed unemployment insurance. In total, about 25,000 workers are in the 1% sample.

In most years since the late 1950's, the price of imported steel has been below the domestic price. This low import price has been a recurrent worry to steel workers and producers, who fear that imports will force domestic mills out of business. Little help is expected from buyer loyalty to American steel: steel is steel and the buyer need only find the cheapest source. The concerns of the steel producers and workers seem, at first, well founded in economic theory: two products which substitute perfectly in use simply cannot sell at different prices; either the higher price will fall to meet the competition or production of the higher priced one will stop. But the production of domestic steel has not stopped, even though domestic producers have lost some ground to imports. Since the late 1950's, when imports first became low priced, the demand for imports has grown steadily and, sometimes, suddenly, as in 1968. Still, the amazing story is not that imports have grown swiftly, but that they have grown slowly. In 1962, imports had 6 percent of the U.S. market. In 1976, 14 years later, they had 14 percent. So each year, on the average, imports have increased their share of the market by less than one percent. (Author)

With few exceptions, occupational licensing has been viewed by economists as a conspiracy of suppliers; they have focused attention almost exclusively on its costs, such as higher prices and restricted opportunities to practice. In this study, an attempt is made to define and test for benefits that might flow from dental licensing. In addition, hypotheses about the determinants of licensing stringency are examined and effects of licensing on the distribution of dentists are investigated. The benefits consumers might realize from the licensing of dentists are: (1) reduced risk of adverse outcomes, (2) reduced cost of acquiring information, and (3) greater satisfaction. Whether these benefits are in fact realized by consumers is subjected to several empirical tests. These tests are based on variation between states in the stringency of licensing. Regression analysis is used to distinguish the effects of licensing from other factors. The study concludes that dental licensing may have certain beneficial effects, and that variation in licensing strengency does not appear to create problems of unequal access to care. Since costs are not estimated, however, or even all benefits, judgment as to whether consumers achieve higher levels of welfare because of licensing must await further research. (Author)

The Federal government tries to keep compensation of Ifs civilian employees competitive with compensation in the private sector by setting wages to be comparable to wages in the private sector. Reliance on wages to measure the comparability of compensation may be inferior to reliance on other measures that incorporate pecuniary and non-pecuniary factors not measured by wages alone. Theory suggests that the quit rate could be used to indicate when compensation is getting out of proper adjustment: a rise in the quit rate would indicate that total compensation is falling in comparison with compensation offered by other employers. To judge whether the quit rate is a sensitive measure of compensation comparability, the relation of quits and relative wages in manufacturing industries was examined. Two different data sets were used to test the same general model. Both tests showed that quits rise when relative wages fall, and vice-versa. In one test, using aggregate time series data for each of twenty-six manufacturing industries, all but two industries showed this negative relationship, and the results were statistically significant in twelve of the twenty-six. The conclusion is that the quit rate is a sensitive measure of wage comparability. If it is to be used as a basis for adjusting federal wages, however, more must be known about hiring costs and about how quits in federal employment should be measured.

In this paper, we describe an improved method of measuring the private and social benefits and costs associated with speed and give some simplified numerical examples. The examples illustrate that, without arbitrarily judging the value of human life, it is possible to: (1) estimate the optimum speed limit; (2) estimate the cost, per life saved, of a speed limit below the optimum, which can be compared with the cost of saving lives in other ways; and (3) specify the types of information needed to improve estimates of the optimum. We have considered only the simplest case, in which all drivers are rational and informed. Further, drivers have the same parameters and, hence, would choose to go the same speed if there were no speed limit. Thus, we are abstracting from any variation in speed and the externalities that fast and slow drivers impose on each other. Our method involves using the speed individuals would go if unencumbered by a speed limit. For this, we need to observe how fast drivers go in places where a limit is absent. The optimum speed limit is then obtained by adjusting this private speed so that the marginal benefits of extra speed equal the social marginal costs rather than the private. We conclude on the basis of our analysis that even if everyone would go the same speed without a speed limit, there are externalites present that justify a speed limit; and that the range of uncertainity about the optimal speed limit can be reduced drastically by paying close attention to driver behavior when there is no speed limit., Revision to report dated May 81, AD-A100 425.

This report contains Appendices to PP 396: Older Workers in the Market for Part-Time Employment. Contents include: A general equilibrium theory of hours and compensation per hour; Evidence of the part-time penalty; An empirical inquiry into the relationship between hours worked and compensation; Policy analysis; and The age compensation of different industries., Appendices to report dated Aug 83, AD-A133 655.

This paper examines the sources and policy implications of the sparsity of part-time work among older workers. One starting point is the contradiction between expectation -- that the transition out of the labor force will be gradual by means of part-time work -- and what appears to be the reality -- that retirement is sudden. Another starting point is the perception of older workers that part-time work at their current wage is scarce. We begin by reviewing evidence from a number of data sets indicating that, despite statements by older workers that they have a strong interest in part-time work, in most cases retirement is sudden. The country's limited number of partial retirees are concentrated in a few industries (agriculture, service, finance/insurance/real estate). Partial retirement is rare in manufacturing, public utilities, and government. Workers approaching retirement age are not spread evenly across industries. Construction, transportation, and finance/insurance/real estate have a high-than-average proportion. The industries that retain the highest proportion of older workers are trade and government. The industries that accept older workers most readily from other industries are agriculture and trade.

This paper presents estimates of what effect technical change had on labor demand from 1958 to 1977 in two important U.S. industries --steel and autos. Both of these industries have, over the period studied, experienced technological innovation; new methods of production have been developed and introduced. At the same time, their employment experience has been mixed. Employment grew over the time period in the auto industry, but fell in the steel industry. This kind of mixed pattern makes it difficult to relate technical change and employment. Technical change might have decreased employment by displacing workers with new machines and equipment, or it might have increased employment by helping to keep these industries competitive in world markets.

Unemployment insurance (UI) is intended to help people who are temporarily unemployed and seeking suitable employment. It is not intended for individuals who are unable or unwilling to work. This paper examines the degree to which screening reduces unemployment and whether the screening procedures used are appropriately targeted on individuals who have little interest in returning to work. The primary focus is on the work test which is a form of screening designed to determine if an otherwise qualified individual is actively seeking suitable work. Two effects of work test screening are examined by studying denial of benefits: the direct effect of denying benefits to those found not to be searching and the indirect or deterrent effect on the claimant population as a whole. Separate models are estimated to examine each effect using and unusually detailed microdata base for a sample of 700 claimants in Arizona during the period 1967-69. The key conclusion of the study is that the threat of a severe penalty i.e., the long term disqualification of a claimant, encourages a substantially faster return to, at least, part-time work. This finding explains why earlier, more aggregative, studies showed that small increases in denial rates were correlated with large reductions in unemployment. In addition, we concluded that a mild penalty, i.e., a disqualification with no loss of benefit entitlement, was too weak to have measurable effects.

It is difficult to find reliable industrial and occupational histories of individuals who have been identified as suffering from particular diseases, such as cancer. State Employment Security Agencies (SESA) maintain records of employment of all workers covered by Unemployment Insurance. To determine the availability of useful data maintained by these agencies, SESA in twelve states were visited and the characteristics and availability of archived data were verified. A telephone survey was also conducted to determine the characteristics and availability of death certificates and other health data in those states. The period and form of storage of archives varied from state to state; Texas and New Mexico maintain data on microfilm going back more than forty years. Colorado has records on computer tape going back to 1968. Computer tape storage in other states covers periods ranging from two to ten years. Details about the characteristics and availability of matchable health and unemployment data from twelve states are presented in this report.

Community-acquired bacterial meningitis in Vermont children under 5 years of age was recognized less frequently in 1967-1970 in those towns with low total hospitalization rates than in towns with hospitalization rates above 15 admissions per 100 population. Using the towns with high hospitalization rates as a norm, it was found that towns with fewer recognized meningitis cases than expected had significantly greater rates of death from obscure causes in children 1-59 months of age in 1967-1970. It is suggested that about 17 deaths in 1967-1970 in Vermont children 1-59 months of age were associated with the failure to recognize bacterial meningitis in children from towns with low rates of medical care utilization.