Motor carrier deregulation and highway safety: an empirical analysis.

• Author: Alexander, Donald L.

1. Introduction

   The passage of the Motor Carrier Reform and Modernization Act of 1980 (MCA 1980) effectively marked the end of rate and entry regulation in the interstate trucking industry, and most economists agree that this legislation has created many important economic benefits during the 1980s. For example, shipping rates fell in real terms, service expanded and improved in quality, and generally resources were used more efficiently.(1) There is increasing concern, however, that deregulation has led firms to reduce safety expenditures in order to remain competitive, and that trucking accidents and fatalities have increased as a consequence [4; 9]. Indeed, Brock Adams (former U.S. Secretary of Transportation) claims that trucking accidents have increased since 1980 for this very reason.(2)

   The limited empirical evidence reported in the literature suggests otherwise. Moore [17], for example, shows that accident, fatality, and injury rates have fallen since 1980 despite the rapid increase in the number of truck-miles traveled [13]. His analysis, however, is based on a comparison of rates across time, and does not reveal any of the potential economic or institutional forces that may be affecting the evolution of these data. in a more systematic analysis, Traynor [22] finds that deregulation has reduced accident rates in California, although it may be difficult to extend his results to the national experience as a whole.

   This paper has two major objectives. The first is to estimate an empirical model using a pooled, cross section of state data to determine whether the evidence Traynor [22] reports for California holds, in general, for all other states. The second objective is to determine empirically those factors that may be driving the results in Moore [17]. Pooling these data allow us to test for the impact of deregulation on accident, fatality, and injury rates while holding state-specific factors constant; an empirical approach that is not possible at a higher level of aggregation.(3)

   The paper is organized in the following manner. In the second section I discuss three possible ways deregulation may have affected accident rates in interstate trucking.(4) The third section presents the empirical model and regression results, and the final section summarizes the major conclusions drawn from this investigation.

2. Motor Carrier Deregulation and Highway Safety

   The current theoretical literature provides a framework for discussing the potential link between changes in deregulation and highway safety.(5) I will use this framework to focus on three aspects of deregulation that are likely to affect highway safety in the trucking industry.

   Before deregulation, rate bureaus acted as cartels and set shipping rates at supracompetitive levels.(6) Since the Interstate Commerce Commission (ICC) restricted the entry of new trucking firms, incumbent firms were able to earn economic rents that were shared with organized labor.(7) After deregulation, however, the ICC essentially permitted free entry and naturally there was an influx of new firms. Winston et al. [31], for example, report that the number of truckers with ICC operating authority increased from 18,045 in 1980 to 36,948 in 1986.[8] One implication for highway safety is that trucking accidents may have increased because of the additional traffic congestion. Moreover, if the new entrants were inexperienced in handling trucks on congested highways, it is likely accidents would have increased until the new drivers gained the necessary driving experience.(9)

   The implicit assumption underlying this argument is that a driver's behavior towards safety remains unchanged in response to a change in the regulatory environment, which Peltzman [18] has questioned on theoretical grounds.(10) Although deregulation in the trucking industry did not involve a specific change in safety regulation, Peltzman's model provides a useful framework to examine the effect of deregulation on the behavior of owner-operators.(11)

   The basic model is that drivers face a choice between driving intensity (e.g., faster speeds, longer hours) and the probability of an accident, which is affected by several economic factors: the price of an accident, income, and a regulatory parameter which we will associate with the driver's promotion of safety. In a cross-section model, it is difficult to argue that interstate drivers respond to differential "accident prices" across each state. Moreover, it is equally difficult to argue that secular changes in income that Peltzman discusses are at work to the same extent in this analysis since we are using data for a shorter time span. Therefore, I will ignore the first and second factors and focus on the third.

   We can think of deregulation shifting the driver's demand for driving intensity in several ways. The widely-held view is that deregulation has led owner-operators to reduce safety expenditures to remain competitive, which raises the probability of an accident for a given level of driving intensity. It is not quite clear, however, why owners would reduce safety expenditures when their discounted future profits depend on providing timely and safe deliveries today.

   On the other hand, deregulation may have induced firms, which employ drivers for hire, to increase their safety expenditures." Suppose a trucking firm uses two complementary inputs, labor (L) and safety (S), per truck to deliver a given quantity of goods. We can think of S broadly as the resources the firm uses to maintain some level of safe operation, which may include any investment in safety training for drivers or the purchase of truck-related safety equipment. In competitive markets, the level of safety the firm provides determines the wages the firm must pay to compensate drivers for any relative risks in trucking.(15) If, for example, the firm offers better training or installs more (or better) safety equipment, then wages would be commensurately lower. As wages fell after deregulation, this may have created an incentive for firms to hire more drivers and to provide additional safety to compensate drivers for any apparent risks in trucking. In addition, it is quite possible that firms provided the drivers with greater safety resources (i.e., equipment or training) because the new drivers were relatively inexperienced. The implication is that if firms increased safety expenditures because wages fell, then accidents would have probably fallen as well. Moreover, if the additional expenditures were used to purchase truck-related safety equipment, then fatalities and injuries would have fallen too.

   And finally, deregulation may have affected driving intensity directly if drivers were induced to make more deliveries by driving longer hours and at faster speeds. Therefore, it would be important to control for differences in speed limit enforcement and vehicle inspections when attempting to explain differences in accident rates across states.

   The above discussion indicates there are reasons to expect that accidents may have increased or decreased as the result of deregulation. The next section discusses the empirical model which attempts to determine the net impact of deregulation, while...