Search

State medical marijuana programs have proliferated in the United States in recent years. Marijuana sales are now estimated in billions of dollars per year with over two million patients, yet it remains unlawful under Federal law, and there is limited and conflicting evidence about potential effects on society. We present new evidence about potential effects on crime by estimating an economic crime model following the general approach developed by Becker. Data from 11 states in the Western United States are used to estimate the model and test hypotheses about potential effects on rates of violent and property crime. Fixed effects methods are applied to control for state-specific factors, with adjustments for first-order autocorrelation and cross-section heteroskedasticity. There is no evidence of negative spillover effects from medical marijuana laws (MMLs) on violent or property crime. Instead, we find significant drops in rates of violent crime associated with state MMLs.

This article provides new evidence concerning the effect on private investment of allocating resources to public consumption and investment. An autoregressive distributed lag model developed for cointegration-error correction analysis is estimated using data for the U.S. public sector for 1956Q1-2010Q2. The most important finding is that in the long run, there is no crowding out associated with the net effect of equal percentage changes in government purchases of domestic consumption and investment. Contrary to some previous results, public investment has a significant crowding in effect on private investment while military purchases have a significant crowding out effect, with an elasticity of -0.52. The evidence also indicates that there is an important investment role for profit growth, but not for the real interest rate in the long run. The results are generally consistent with post Keynesian views of fiscal policy and support those who argue that the 2009 stimulus package was not well-suited for gene...

Explanations for the Great Moderation in GDP volatility have included improved management of inventory schedules, the good luck of smaller economic shocks, and better anti-inflation policy. This article provides direct evidence on the changes in production behavior underlying these explanations within a market model for consumer durable goods. Long-run price and sales elasticities are estimated using VECMs for 1959 through 1983 (period I) and 1984 through 2008 (period II). Significant and more effective adjustments to output growth in response to both market disequilibria and changes in demand occur in period II and contribute to the reduced volatility observed. During that time, 95 percent of market disequilibrium gaps were closed after four quarters, and current output adjusted to accommodate 90 percent of demand changes occurring during the preceding three quarters.

Using data from 1955 to 2005, the estimates presented imply a long-run equilibrium relation between U.S. federal government revenues and expenditures, but one that is consistent with a continuous expansion in the deficit. There is evidence of a structural break that led to deficit reduction and surpluses from 1996 to 2001, which coincides with several noteworthy economic and political outcomes including the only time when a Democratic administration and Republican Congress were in power simultaneously. The results support the view of a long-run deficit bias in federal budget decision-making. Absent behavior similar to that of the late 1990s, a change in budget rules may be needed to avoid the large deficits forecasted under current law.

U.S. law enforcement against the sale and possession of marijuana has been estimated to cost close to $8 billion a year in criminal justice resources. Current enforcement is justified if it provides net benefits greater than alternatives such as a legal, regulated market for marijuana. Prior research suggests that current levels of drug enforcement may increase nondrug crime and hard drug use. Here, local rates of property crime, homicide, and nonmarijuana drug possession are estimated as a function of economic conditions, enforcement effectiveness, and arrests for possession or sale of marijuana. The data consists of a pooled sample of over 1300 U.S. counties (1994–2001). The results suggest that marijuana arrests are associated with increases in homicides, burglaries, motor vehicle thefts, and larcenies along with subsequent increases in hard drug arrests. These results raise significant questions about the merits of policies that focus on criminal justice approaches to marijuana control.

Summary and Conclusions A growing body of research uses the economic model of crime to evaluatethe role of enforcement, economic conditions, and other characteristics oflocal crime markets. This analysis of the determinants of county-level crime 16 Benson, Kim, and Rasmussen (1998) provided theoretical arguments for why arrestvariables should be treated as exogenous, and found empirical support for this specificationusing econometric tests for exogeneity. They also maintained that fixed-effects models reducethe potential of endogeneity problems for independent variables included in the regressionequation. Miron (1999) reached a similar exogeneity conclusion because the level of policeresources will be related to lagged, not current, crime rates due to the political processesassociated with changing the allocation of police resources. 17 There is inconclusive evidence about the relationship between drug use and crime, andalthough significant correlations are clearly present, strong support is lacking for the hypoth-esis that drug use causes crime. Rasmussen and Benson (1994) provided a comprehensivereview of the evidence and concluded that, except for a small subset of drug users, drug useand crime appear to be unrelated. Miron (2003:16) argued that ‘‘the evidence . . . demon-strates a correlation between a tendency to commit crime and the tendency to use drugs,without indicating whether there is a causal connection,’’ and that ‘‘reviews of the literature ondrug use and crime have consistently concluded there is little evidence that drug use per secauses crime.’’ Mast, Benson, and Rasmussen (2000:292) also found that ‘‘substantial researchliterature suggests that there is no reliable association between drug use and major non-drugcrimes.’’

This paper evaluates the relationships among federal anti-drug law enforcement expenditures, education and treatment expenditures, and public health outcomes. The data include four types of spending: criminal justice system, interdiction and international intelligence, education in the community and workplace, and drug treatment. These data were combined with mortality rates for drug abuse, a public health outcome. The empirical findings support the hypothesis that resources allocated to drug prevention and treatment have benefited the public health. Conversely, a 10% reduction in enforcement expenditures is associated with a long-run reduction of approximately 3,000 deaths per year.

A demand-based model of public sector spending is used to identify the economic determinants of the relative size of government in the United States for the period 1949 to 1998. The government's share of GDP is hypothesized to be a function of income growth (Wagner's Law), relative prices for public sector output (Baumol/Beck hypothesis), and several demographic, cyclical, and budget variables. Although nondefense, noninterest government spending as a share of GDP has not grown in recent decades, the results support both income and price explanations of increasing government size. It is shown that the income and price effects are sensitive to the degree of disaggregation used to define government spending shares. Higher relative prices outweigh economic growth as a factor behind increasing shares for government consumption. Economic growth dominates the relative price effect for investment purchases. The two effects are similar for transfer payments.

Capital stock data for the U.S. economy are used to develop a measure of sectoral shifts in productive resources. Within the context of the creative destruction process, this measure provides a direct indicator of sectoral shifts in resource demands independent of aggregate fluctuations. Years with greater reallocations of capital have higher unemployment, a result consistent with the traditional sectoral shifts hypothesis. However, fluctuations in unemployment appear to be more strongly influenced by aggregate rather than sectoral shocks. Significant variation exists across demographic groups in the responsiveness of unemployment to aggregate fluctuations and sectoral shifts. The adverse impact of sectoral shifts is greater for males and members of the nonwhite labor force.

A model of macroeconomic complementarity is used to assess causes of comovement in investment spending across nine sectors of the U.S. economy. It is hypothesized that the irreversibility and uncertainty of investment spending imply a greater role for investment linkages and aggregate factors in investment fluctuations compared with estimates for employment and output. For the average sector, past investment growth across all sectors, changes in aggregate demand, and a common factor account for two-thirds of the variance of investment growth. After accounting for aggregate demand, sectoral shocks explain 70% of the average sector’s innovations to investment growth.

This article tests for cointegration between unit labor costs and the level of product prices in four sectors of the U.S. economy: the aggregate business sector, the nonfinancial corporate sector, durable manufacturing, and nondurable manufacturing. A finding of cointegration for most specifications supports the existence of long-run labor market equilibrium for producers and suggests estimation of error-correction models to examine the dynamic relationships. In every sector except nondurable manufacturing, error-correction model estimates indicate there is a mutual feedback relationship between unit labor costs and prices. Controlling for deviations from full employment, oil price shocks, and the Nixon wage and price controls, the results also provide evidence of significant nominal wage indexation in U.S. labor markets. Throughout the economy there appears to exist both effective neoclassical wage and price adjustment mechanisms to maintain labor market equilibrium and short-run rigidities which may contribute to deviations from full employment outcomes.

Cross-sectional data are used to assess the effect of state-level economic conditions on state outcomes in the 1992 presidential election. The analysis provides evidence on the role of macroeconomic variables in models of national election outcomes and presents simulations to determine whether changes in economic circumstances might have reversed the election's result. It is argued that self-interested voters are more likely to prefer a new president if they are experiencing unemployment or income losses, or if they fear that economic conditions may lead to their own unemployment or inadequate income growth. Altruistic voters may seek a change to improve the economic positions of others. Whether specified as the current unemployment rate or as recent real per capita income growth, states with subpar economic performances are found to have significantly higher voter shares for Clinton. The simulations indicate that reasonable adjustments in state unemploy ment rates do not reverse the election outcome, whereas more rapid personal income growth during the year prior to the election results in a Bush victory.

The Office of Management and Budget (OMB) annually submits the President's budget for the U.S. government to the Congress. The economic forecasts and revenue and outlay proposals contained in the budget have been criticized as biased, especially during the 1980s. Tests for bias in one year ahead proposals for the 1963–89 period show no bias in economic forecasts and revenue estimates, but substantial bias in outlay proposals. Most outlay proposals are consistently less than actual outcomes, accounting for underprediction of the Federal deficit. OMB outlay proposals appear to be influenced by politics. Republican administrations show more significant proposal biases, with defense proposals higher and domestic outlay proposals lower than outcomes. The Office of Management and Budget consistently understates the deficits by resorting to the most optimistic economic assumptions it can credibly — and now sometimes even incredibly — employ. — Senator James SasserWall Street Journal, 25 January 1990

Some analysts blame the large state discretionary revenue increases of the early 1990's on recessionary declines in revenue and on federally mandated spending increases. Others cite rapid increases in state government spending, particularly on employee compensation, during the 1980's. Test the determinants of state discretionary revenue increases during the 1991 and 1992 fiscal years using regression analysis and finds both explanations to be equally important.

Intervention analysis is used to assess the impact of the October 1987 stock market crash upon employment in the securities industry in New York City. The results indicate that the crash has led to a gradual, but permanent reduction of approximately 25,000 jobs in this industry through July 1990. Using a dynamic location quotient methodology to measure basic employment, an estimate of a partial-adjustment economic base multiplier suggests that total New York City employment is likely to decline by over 100,000 as a result of the crash and subsequent adjustments in the securities industry.

Okun's Law is estimated for 26 states in the United States. An average of 3.1 percentage points of additional state-level output growth is found to be required for each percentage point reduction in a state's unemployment rate. Interstate differences in industrial mix, the age and gender distributions of the labor force, the rate of labor force growth over the sample period, and tax policy decisions are significantly related to spatial variation in Okun's Law. The results lend support to and extend previous explanations of the value of Okun's Law for the national economy.

Static and dynamic translog cost models for the aggregate U.S. state and local government sector are estimated for 1952 to 1985. Inputs are capital, labor, and a combined nondurable goods and other services input, which includes "privatized" services. Estimates show that capital and labor are complements and own-price inelastic. The nondurable-service input is a substitute for both labor and capital and has an own-price elasticity closer to unity. It is used relatively more as the sector expands. The results offer an explanation for the characterization of the government as a low productivity industry and show an economic rather than political motivation for the privatization of some public services.

A translog cost model is applied to three military inputs: labor, equipment, and structures for the United States, 1929–87. The results show labor and equipment to be substitutes, with a decreasing elasticity of substitution since 1929, perhaps due to decreasing flexibility in defense technology. Since 1964 defense quantity increases have favored the stockpiling of equipment. Both the Korean and Vietnam wars caused a relative increase in the use of labor. The end of the draft of 1973 shifted input use away from labor, towards equipment.

Monthly data for the aggregate U.S. economy are used to assess competing hypotheses concerning the relationship between sectoral employment shifts and fluctuations in the unemployment rate. It is shown that sectoral shifts are caused by major work stoppages, aggregate fluctuations unrelated to permanent sectoral shifts, a reallocation-timing effect, and allocative shocks at the sectoral level. Larger employment shifts are associated with higher unemployment during slow growth periods and lower unemployment during above-average growth periods. Models are presented which demonstrate that both aggregate and allocative shocks are causes of cyclical increases in unemployment.

A model of urban manufacturing location is developed in order to explain the demand for industrial sites in a metropolitan area. Within the theory of qualitative choice behavior, firms identified by a set of eight characteristics are viewed as selecting sites possessing two attributes of varying degrees of importance to different kinds of firms. Using a multinomial logit specification, the model is estimated for the Cincinnati SMSA. The results lend support to the firm-specific nature of the site selection process. Information concerning the determination of the spatial distribution of the firm types in Cincinnati is derived from the analysis. Implications for the possible effectiveness of a policy initiative designed to alter existing patterns of urban industrial location conclude the paper.

Recent legislation has reduced federal tax rates and provided for indexation of the personal income tax against inflation. These changes are in part designed to reduce the relative size of government in the U.S. economy. Testing assump tions behind this proposition, this article examines the causal relation between revenue and expenditure changes in explainingfederal budget growth. Statistical causality tests reveal that revenue growth generally precedes expenditure growth, confirming that growth in revenue capacity stimulates budget expan sion rather than deficit reduction. A trivariate causal model including GNP, however, suggests that both demand and supply factors are responsible for the budget's growth.

A great deal of attention recently has been focused on issues related to spatial differentials in labor productivity and the external economies of size.' Historically size economies have been related to scale economies and plant capacity.2 More recently industry and city size externalities, often termed localization and urbanization economies or simply agglomeration economies, have been analyzed to determine the direction and extent to which these size factors affect labor productivity and economies of scale [15]. A variety of issues characterize this literature, among which are the magnitude and direction of these effects, the appropriate estimating equation, and the relation between size, returns to scale, and the neutrality or factor bias of these effects. Despite their apparent success in identifying the relation between various size and productivity measures, past studies are not without their shortcomings. The purpose here is to address these issues and shortcomings in the context of estimating the external effects of industry size on labor productivity. Along the way we will indicate the shortcomings of approaches recently employed and consider the empirical importance of different measures of industry size and capital.' The next section reviews some of the shortcomings of past studies. They are of particular interest, for they are avoided by the approach and data of this study. The analytical approach is formulated in Section II which includes a report of the empirical analysis and its major findings. The final section summarizes and discusses the implications of this analysis.

A Rotterdam demand framework is used to estimate the demand for five state and local government (SLG) services. A prefere nce independence transformation of these results identifies the under lying voter wants which these services satisfy. The results reflect a shift from physical capital to human capital wants over the 1952-83 period. A general SLG want accounts for 75 to 95 percent of SLG spend ing over this period. A "minimum service level" want, including mos t highway and infrastructure spending, declines in importance over ti me, while a "human capital/infrastructure contrast" want, composed of education and health spending, increases in importance. Copyright 1987 by MIT Press.