A radical change in traffic law: effects on fatalities in the Czech Republic
01 Dec 2014-Journal of Public Health (Oxford University Press)-Vol. 36, Iss: 4, pp 539-545
TL;DR: In this article, the authors examined short-and long-run effects of a new-stricter road traffic law on traffic accident-related fatalities in the Czech Republic and found that there was a sharp, 33.3% decrease in accidentrelated fatalities during the first three post-reform months.
Abstract: BACKGROUND: This study examines short- and long-run effects of a new-stricter-road traffic law on traffic accident-related fatalities in the Czech Republic. The law introduced tougher punishments through the introduction of a demerit point system and a manifold increase in fines, together with augmented authority of traffic police. METHODS: Identification is based on difference-in-differences methodology, with neighbouring countries serving as a control group. RESULTS: There was a sharp, 33.3%, decrease in accident-related fatalities during the first three post-reform months. This translates into 127 saved lives (95% confidence interval: 51, 204). The decline was, however, temporary; the estimates of the effects going beyond the first year are around zero. Unique data on traffic police activity reveal that police resources devoted to traffic law enforcement gradually declined. CONCLUSIONS: Tougher penalties have significant, but often short-lived effects. Weaker enforcement in the aftermath of such reforms may explain the absence of long-run effects. Language: en
Summary (5 min read)
Jump to: [1 Introduction] – [2 The Change in Czech Road Traffic Law] – [3.1.1 Data Collection] – [3.1.2 Fatalities and Injuries] – [A: Totals by period before and after July 1, 2006] – [3.1.3 Transport Statistics and the Economy] – [3.2.1 The Outcome of Interest] – [3.2.2 Empirical Model] – [3.2.3 Validity of Assumptions] – [3.2.4 Meaning of Estimates] – [4.1 Main Results] – [4.2 Short-Run Development] – [5.1 Competing Explanations and Some Theory] – [5.2 The Data on Police Activity] – [5.3 What Does (Not) the Traffic Police Activity Explain?] and [6 Conclusions]
- Each year, road traffic accidents (RTAs) result in as many as 50 million injuries and more than 1.2 million deaths, making it the ninth leading cause of death worldwide— effecting especially young people—and its importance is predicted to rise over the next two decades.
- This study evaluates the effects of a new road traffic law in the Czech Republic that became effective on July 1, 2006 (Parliamant of the Czech Republic 2005).
- The effects going beyond the initial six months are ambiguous, as many of these studies are based on short-term data and there are contradictions among those that do look at long-run effects.
- To the extent that the development of the variable of interest is similar across these countries, the control group allows estimating the counterfactual, i.e. the hypothetical scenario of what would have happened on Czech roads had the law not been enacted.
2 The Change in Czech Road Traffic Law
- The mechanics of the newly introduced DPS is straightforward and relatively strict.
- A driver who accumulates 12 points has his license revoked for 12 months, automatically and immediately.
- Maximum fines for offenses that can be solved on the spot, if driver accepts the ticket, were mostly raised twofold, but fines for speeding were tripled, as illustrated in Figure 1.10 Similarly, fines were increased for more serious offenses that are dealt with by the municipal office.
- Police regained the capacity to retain a driver’s license on the spot and if a driver refuses an alcohol test, the police can seize his vehicle or prevent the driver from continuing.
3.1.1 Data Collection
- The main data set analyzed in this paper consists of monthly regional-level data on RTA-related casualties that occurred between January 2004 and December 2008 in the Czech Republic, Germany, and Austria, obtained upon specific requests from the Czech Traffic Police Headquarters and statistical offices of Germany and Austria.11.
- I then merged this data with yearly regional-level data on the population and number of cars from Eurostat and yearly country-level data on transport and economic statistics from the same source.
- In addition, I received daily data on fatalities in Austria and the Czech Republic covering 2005 to 2008.
- Finally, from the Czech Traffic Police Headquarters.
- I have obtained detailed information on traffic police activity, such as man-hours, hours of use of speed guns, the number of cleared offenses, and the amount of collected fines.
3.1.2 Fatalities and Injuries
- Table 1 summarizes the data on RTA-related casualties in Austria, the Czech Republic, and Germany (split by former East and West) before and after July 1, 2006.
- Looking at the levels of fatalities, the Czech Republic had the highest rate per million inhabitants as well as per million cars in both periods.
- The picture is quite different when the authors look at injuries, however.
- Czech rates of RTA-related slight and serious injuries per million inhabitants are about 50 percent lower compared to Austria and Germany, while the number of injuries per million cars is still about 1/3 smaller, also known as The difference is substantial.
- The decline in injuries always exceeds the decline in fatalities, most notably for the seriously injured.
A: Totals by period before and after July 1, 2006
- Both factors are likely to increase the pool of accidents and injuries unnoticed by the police.
- Since there are no marginal fatalities, the effect of classification should be stronger in the case of serious injuries than slight injuries.
- To probe things further, I compared the police data with yearly data on road traffic fatalities and injuries from the Institute of Health Information and Statistics of the Czech Republic (IHIS).
3.1.3 Transport Statistics and the Economy
- Table A.3 presents a summary of the available transport statistics and GDP from Eurostat.
- At the same time, the average number of passengers per car was decreasing, in fact canceling out the increase in kilometers driven and resulting in a slight decrease in passenger-km per car in the Czech Republic over the period under study.
- The high intensity of transport traffic after the Czech Republic became a member of the European Union in 2004 is often mentioned as potentially elevating the riskiness of Czech roads.
- Finally, the second half of the 2000s was an era of rapid economic growth in the Czech Republic—real GDP per capita measured in 2005 Euros increased by 32 percent.
- The expectation of the effect of GDP growth on RTA-related fatalities per car is ambiguous; in the short run, one would expect increased traffic to dominate, implying a positive correlation between GDP and fatalities.
3.2.1 The Outcome of Interest
- This suggests that the intensity of road traffic was changing and grew faster in the Czech Republic relative to the two neighbors.
- An issue with using the number of cars to adjust for changes in traffic intensity can be that the number of kilometers per car or the average number of passengers sitting in a car may change.
- Both variables are very likely positively related to fatalities.
- 21This reasoning abstracts from the indirect effect of a higher number of car-kilometers on the probability of an accident working simply through more cars meeting on the road.
3.2.2 Empirical Model
- This paper employs a difference-in-differences (DD) estimator using neighboring countries as a control group.
- Tt consists of dummies for four post-reform quarters and a dummy for the fifth to tenth quarter after the reform.
- The parameter of interest is the vector β, which consists of five coefficients capturing the effects of the traffic law reform on fatalities over time.
- The well-known advantage of this estimator is that it helps with controlling for any unobserved shocks, as long as they affect the treated as well as the control group.
- The identifying assumptions of the DD estimator are common trends between the treated and the control group as well as the absence of any unobserved shock specific only to the control countries or to the treated country.
3.2.3 Validity of Assumptions
- There are good reasons to expect that factors generating shocks to RTA-related fatalities are shared among the Czech Republic, Austria, and Germany, so the neighboring countries offer themselves as natural control group.
- I did not receive any response from the German ministry, however researching publicly available resources did not result in finding any substantial law or policy change in Germany between 2004 and 2008.
- 14 of log monthly fatalities per million cars between the Czech Republic and either control country from January 2004 until June 2006 are above 0.7 and are statistically significant at any conceivable level.
- This data include Austria, the Czech Republic, former West Germany, Poland, and Slovakia.
- 24While data are available from 1996, there are frequent revisions, especially in statistics on number of cars, making the earlier data less reliable.
3.2.4 Meaning of Estimates
- In their case, the coefficients β from regression 1 represent time-varying treatment effects on the treated, as the policy change was designed and introduced by the Czech Republic itself, unlike in a random assignment.
- There can also be a potential endogeneity bias in the sense that the timing of the reform may not be random.
- Everyone was long aware that the change in the road traffic law was coming soon and this may have affected pre-reform outcomes, in fact generating some effects of the reform before it actually took place.
- They may also have expected the police to increase their effort around the introduction of the reform.
- 16 the longer the data for the pre-reform period.
4.1 Main Results
- The main set of ordinary least squares estimates of the effects of the new Czech road traffic law on fatalities using regression (1) are reported in Table 2.28 Specification (1) (I will refer to it as the “base result”) shows that the immediate effects of the law were substantial, but shortlived.
- To summarize, these findings provide strong evidence of substantial immediate effects of the new road traffic law on fatalities in the Czech Republic, but not much is apparent beyond that.
- Table 3 probes their base results with additional control variables.
- I first include GDP per capita, which grew faster in the Czech Republic compared to Austria and Germany, as seen in the last column of Table A.3.
- The estimates of long-run effects, that is beyond the first 12 post-reform months, are substantively and statistically insignificant.
4.2 Short-Run Development
- The availability of daily country-level data on RTA-related fatalities in Austria and the Czech Republic makes it possible to study the response tho the new law in more detail.
- There is no apparent positive or negative trend during the pre-reform period, which is reassuring.
- Importantly, the figure corroborates their main finding that the effects of the law were concentrated in the first three months, as most of the estimates for that period lie below the zero line.
- Estimates are less erratic compared to Panel A, as the restrictive model is likely to filter out some noise.
5.1 Competing Explanations and Some Theory
- The short life of the initial effects of the traffic law reform is not too surprising.
- Intuitively, as higher punishments improve drivers’ behavior, enforcement resources may be reallocated to more valuable uses.
- The police dislikes offenses as well as effort.
- There are various reasons why this simple model may capture some of the reality.
- It is also plausible that when similar large-scale changes in the law are adopted, the traffic police may already be overstretched.
5.2 The Data on Police Activity
- This subsection presents an analysis of a unique dataset parsed from internal monthly regional-level information on traffic police activity in 2006 and 2007 I have obtained 40See also Holler (1993) and Andreozzi (2002) for a further discussion of Tsebelis’ model.
- The number of policemen assigned to enforcement a exhibits general upward pattern over 2006 and 2007.
- 25 Tables 4 and 5 study traffic police activity in more detail.
- I use half-yearly country-level data on man-hours in enforcement ranging from 2005 to 2008 and estimate a coefficient for a second half-year, which I then use to deseason the monthly data on police activity.
- Nonetheless, the total amount of traffic police man-hours dedicated to enforcement was gradually declining from the third quarter of 2006 onwards.
5.3 What Does (Not) the Traffic Police Activity Explain?
- In sum, the police data reveal that enforcement levels were declining in the aftermath of the traffic law reform and suggest that traffic police may have put a higher share of resources into general law enforcement and police work and away from the direct enforcement of traffic rules.
- This development is in line with the outlined theory, however one should be careful before interpreting this result strictly causally.
- Only two years of data are available, so the authors cannot rule out preexisting trends or factors driving the changes in police data other than the change in the traffic law.
- It can hardly do so with respect to the sharp short-run decline in fatalities as the police presence on the roads was apparently lower during the third quarter of 2006, relative to first half of that year.
- Doing this would lead to results lacking proper interpretation.
- This study evaluates the effects of the introduction of a new road traffic law on RTA-related fatalities in the Czech Republic.
- Consistent with the literature, studying the effects of similar changes in traffic laws in other countries, I find a substantial initial response to the increase in punishments for traffic offenses.
- From 51 to 204—with point estimate of 119—human lives were saved with 95 percent certainty in that period.
- It is noteworthy in this context that traffic intensity, measured by number of cars, kilometers driven, or intensity of freight transport, was increasing substantially during the period under study.
- A continuous decline in traffic law enforcement cannot, however, explain the short-run pattern of the reaction to the legal reform.
Did you find this useful? Give us your feedback
Center for Economic Research and Graduate Education
Academy of Sciences of the Czech Republic
A RADICAL CHANGE IN TRAFFIC LAW:
EFFECTS ON FATALITIES IN
THE CZECH REPUBLIC
WORKING PAPER SERIES (ISSN 1211-3298)
Working Paper Series 484
A Radical Change in Traffic Law:
Effects on Fatalities in the Czech Republic
Prague, March 2013
ISBN 978-80-7343-288-1 (Univerzita Karlova. Centrum pro ekonomický výzkum
a doktorské studium)
ISBN 978-80-7344-280-4 (Národohospodářský ústav AV ČR, v.v.i.)
A Radical Change in Traﬃc Law: Eﬀects
on Fatalities in the Czech Republic
Faculty of Law, Masaryk University and CERGE–EI
I evaluate the eﬀects of a new road traﬃc law in the Czech Republic that became eﬀective
on July 1, 2006. The law introduced tougher punishments through the introduction of
a demerit point system and a manifold increase in ﬁnes, together with an augmented
authority of traﬃc police. I ﬁnd a sharp, 33.3 percent, decrease in accident-related fatalities
during the ﬁrst three post-reform months. This translates into 51 to 204 saved lives with
95 percent certainty. The decline was, however, temporary; estimates of the eﬀects going
beyond the ﬁrst year are around zero. Unique data on traﬃc police activity reveal that
police resources devoted to traﬃc law enforcement gradually declined and were shifted
towards general law enforcement.
Studie hodnotí efekty Zákona o provozu na pozemních komunikacích v České republice,
který vstoupil v platnost 1. července 2006. Zákon přinesl tvrdší postihy řidičů pomocí
zavedení bodového systému a několikanásobného zvýšení pokut, spolu s posílením pravomocí
policistů ve výkonu služby. V průběhu prvních tří měsíců platnosti zákona nastal prudký,
33,3 procentní, pokles počtu smrtelných zranění. To představuje 51 až 204 zachráněných
životů s jistotou 95 procent. Pokles byl však krátkodobý a odhady efektů přesahujících
první rok jsou kolem nuly. Unikátní data o aktivitě dopravní policie ukazují, že prostředky
pro výkon služby postupně klesaly a byly více alokovány na obecné vynucování práva.
Keywords: Traﬃc law, traﬃc fatalities, policy evaluation, deterrence, enforcement.
JEL classiﬁcation: I12, I18, K42, R41
I would like to thank to Daniele Bondonio, Brendan Dooley, Libor Dušek, Kateřina Holíková,
Michael Kohl, participants at the 2010 Conference on Empirical Legal Studies, the European
Law & Economics Association 2012 Annual Conference, and a workshop at the University of
Economics, Prague for valuable comments and suggestions. I very much appreciate discussions
with oﬃcers at the Czech Traﬃc Police Headquarters as well as their responses to data requests.
The views expressed in this paper as well as any remaining mistakes and imperfections should be
attributed only to the author. This research was funded by Czech Science Foundation grant no.
Center for Economic Research and Graduate Education–Economics Institute, a joint workplace
of Charles University in Prague and the Academy of Sciences of the Czech Republic. Address:
CERGE–EI, P.O. Box 882, Politických vězňů 7, Prague 1, 111 21, Czech Republic.
Each year, road traﬃc accidents (RTAs) result in as many as 50 million injuries and
more than 1.2 million deaths, making it the ninth leading cause of death worldwide—
eﬀecting especially young people—and its importance is predicted to rise over the
next two decades. Enacting comprehensive laws with appropriate penalties and
ensuring necessary resources for enforcement are acknowledged as top instruments
to improve road safety (World Health Organization 2009). However, proper policy
choice requires that we understand how alternative measures perform when put
in place and how they interact with other key variables. Exploiting past policy
experiments is a natural way to improve our understanding of these phenomena.
This study evaluates the eﬀects of a new road traﬃc law in the Czech Republic
that became eﬀective on July 1, 2006 (Parliamant of the Czech Republic 2005).
It was aimed at improving road traﬃc safety through tougher sanctions for traﬃc
oﬀenses and the augmented authority of the police. Apart from a manifold increase
in ﬁnes, the most important change introduced by the law was a demerit point
system (DPS) under which an accumulation of points for traﬃc oﬀenses leads to
the suspension of driver’s license. I overview the new road traﬃc law in detail in
There are over a dozen studies investigating the eﬀects of similar changes in traﬃc
laws that also included a DPS recently adopted in other countries.
these studies in Table A.1 in the Appendix. The common pattern of their ﬁndings
is that the introduction of stricter traﬃc laws is followed by substantial decreases in
RTA-related fatalities and other casualties, usually in the realm of 20 to 30 percent.
However, the eﬀects going beyond the initial six months are ambiguous, as many of
these studies are based on short-term data and there are contradictions among those
that do look at long-run eﬀects. For instance, one study for Ireland ﬁnds lasting
Brazil did so in 1998, Ireland in 2002, Italy in 2003, Spain in 2006, and the United Arab
Emirates in 2008.
The study from the United Arab Emirates is an exception as it does not ﬁnd any eﬀects
TL;DR: In this article, the authors developed and analyzed a simple model of the technical and behavioral mechanisms that determine the volume emissions produced by a car and showed that an optimal fuel tax would provide drivers with appropriate incentives along all three margins and that only public information is needed for a fuel tax to be set optimally.
Abstract: The volume of pollution produced by an automobile is determined by driver's behavior along three margins: (i) vehicle selection, (ii) kilometers driven, and (iii) on-road fuel economy. The first two margins have been studied extensively, however the third has received scant attention. How significant is this 'intensive margin'? What would be the optimal policies when it is taken into account? The paper develops and analyzes a simple model of the technical and behavioral mechanisms that determine the volume emissions produced by a car. The results show that an optimal fuel tax would provide drivers with appropriate incentives along all three margins and that only public information is needed for a fuel tax to be set optimally. In contrast, an optimal distance tax would require private information. Lastly, relative to the optimal fuel tax, a simple uniform fuel tax is shown to be progressive. Thus, being already deployed worldwide, a uniform fuel tax is an attractive second-best policy. These findings should be accounted for when designing new mechanisms to alleviate motor vehicle pollution.
TL;DR: In this article, the authors examined the records of 28 current member states of the European Union over the 1999-2010 period and found that road traffic fatality rates are higher in countries where the behavior of the inhabitants is more aggressive, while the rates are lower in countries with more severe penal systems.
Abstract: The connection between crime and road safety is a relatively recent topic in academic research, although most studies have focused on the link between criminal behavior and traffic offenses, and only a few authors discuss the possible relationship with traffic accident fatalities. Evidence worldwide shows that people who commit other offenses characteristic of antisocial attitudes are more likely to have road traffic accidents and infringe traffic laws. We examine the records of the 28 current member states of the European Union over the 1999-2010 period. Our aim is to test the hypothesis that crime rates (and specifically, motor vehicle-related crimes) can be considered as predictors of fatal road traffic accidents. If they can be, this could be prima facie justification, at least, of the trend in several countries to consider traffic offenses as crimes in their penal codes and to toughen the punishment imposed on those who commit them. The effect of the severity of the legal system applied to traffic offenses is also analyzed. From a geographical point of view, our results reveal that road traffic fatality rates are higher in countries where the behavior of the inhabitants is more aggressive, while the rates are lower in countries with more severe penal systems. Language: en
TL;DR: Increasing traffic fines was found to be associated with small changes in the rate of violations, and a small reduction in fatal accidents was associated with increased fixed penalties, varying between studies from less than 1-12%.
Abstract: Studies that have evaluated the association between increases in traffic fine amounts (fixed penalties) and changes in compliance with road traffic law or the number of accidents are synthesised by means of meta-analysis. The studies were few and different in many respects. Nine studies were included in the meta-analysis of changes in compliance. Four studies were included in the meta-analysis of changes in accidents. Increasing traffic fines was found to be associated with small changes in the rate of violations. The changes were non-linear. For increases up to about 100%, violations were reduced. For larger increases, no reduction in violations was found. A small reduction in fatal accidents was associated with increased fixed penalties, varying between studies from less than 1-12%. The main pattern of changes in violations was similar in the fixed-effects and random-effects models of meta-analysis, meta-regression and when simple (non-weighted) mean values were computed. The main findings are thus robust, although most of the primary studies did not control very well for potentially confounding factors. Summary estimates of changes in violations or accidents should be treated as provisional and do not necessarily reflect causal relationships.
TL;DR: It is concluded that the penalty point system has a significant deterring effect for drivers who are at high risk of losing their license at the next infraction.
Abstract: We analysed data from the Norwegian driver’s licence penalty point register over a three-year period, in order to investigate whether the number of incurred penalty points in a given time period can predict the probability of incurring additional points in the subsequent period. Data for all category B drivers without penalty points at the start of the study period were included in the analyses. Norway’s penalty point system implies that speeding and various other traffic violations result in two or three penalty points for full-license drivers and four or six points for probationary-license drivers. Eight points within a three-year period results in a six-month disqualification. Two hypotheses were formulated: 1) A “driving style effect” implying that drivers with previous penalty points have a higher probability of incurring new points than drivers without previous points; and 2) a “deterrence effect” implying that drivers with more than four points have a reduced probability of incurring new points, due to impending risk of license revocation. Results showed an inverted U-shaped relationship between number of penalty points incurred during a one-year period and the number of additional penalty points incurred in the subsequent year, with the highest number for drivers with four previous points. Thus, both hypotheses were clearly supported, and it is concluded that the penalty point system has a significant deterring effect for drivers who are at high risk of losing their license at the next infraction.
TL;DR: In this article, the authors discuss the links between corruption in the traffic police and road safety and show that corruption has a positive impact on road safety in Russia, a middle-income country.
Abstract: Motorists deal with traffic police officers on a daily basis. In Russia, the operations of the traffic police are not transparent. Mass surveys show that contacts with traffic police officers represent a key source of corruption in this country. This article discusses the links between corruption in the traffic police and road safety. Corruption in the traffic police has a positive impact on road safety in Russia, a middle-income country. It suppresses economic growth and thus reduces the intensity of road use. In the current situation, Russian motorists have no incentive for fighting corruption: constantly growing fines and penalties for traffic offences increase the attractiveness of paying bribes compared to individual and/or collective protests. A vicious circle emerges as a result: corruption becomes self-sustainable. The official statistical data and results of a nationally representative sociological survey provide the data for the analysis. An instrumental variables analysis and multiple regression modelling are used in this study.
TL;DR: Copyright (©) 1999–2012 R Foundation for Statistical Computing; permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and permission notice are preserved on all copies.
Abstract: Copyright (©) 1999–2012 R Foundation for Statistical Computing. Permission is granted to make and distribute verbatim copies of this manual provided the copyright notice and this permission notice are preserved on all copies. Permission is granted to copy and distribute modified versions of this manual under the conditions for verbatim copying, provided that the entire resulting derived work is distributed under the terms of a permission notice identical to this one. Permission is granted to copy and distribute translations of this manual into another language, under the above conditions for modified versions, except that this permission notice may be stated in a translation approved by the R Core Team.
01 Jan 2001
TL;DR: This is the essential companion to Jeffrey Wooldridge's widely-used graduate text Econometric Analysis of Cross Section and Panel Data (MIT Press, 2001).
Abstract: The second edition of this acclaimed graduate text provides a unified treatment of two methods used in contemporary econometric research, cross section and data panel methods. By focusing on assumptions that can be given behavioral content, the book maintains an appropriate level of rigor while emphasizing intuitive thinking. The analysis covers both linear and nonlinear models, including models with dynamics and/or individual heterogeneity. In addition to general estimation frameworks (particular methods of moments and maximum likelihood), specific linear and nonlinear methods are covered in detail, including probit and logit models and their multivariate, Tobit models, models for count data, censored and missing data schemes, causal (or treatment) effects, and duration analysis. Econometric Analysis of Cross Section and Panel Data was the first graduate econometrics text to focus on microeconomic data structures, allowing assumptions to be separated into population and sampling assumptions. This second edition has been substantially updated and revised. Improvements include a broader class of models for missing data problems; more detailed treatment of cluster problems, an important topic for empirical researchers; expanded discussion of "generalized instrumental variables" (GIV) estimation; new coverage (based on the author's own recent research) of inverse probability weighting; a more complete framework for estimating treatment effects with panel data, and a firmly established link between econometric approaches to nonlinear panel data and the "generalized estimating equation" literature popular in statistics and other fields. New attention is given to explaining when particular econometric methods can be applied; the goal is not only to tell readers what does work, but why certain "obvious" procedures do not. The numerous included exercises, both theoretical and computer-based, allow the reader to extend methods covered in the text and discover new insights.
TL;DR: In this article, the authors randomly generate placebo laws in state-level data on female wages from the Current Population Survey and use OLS to compute the DD estimate of its "effect" as well as the standard error of this estimate.
Abstract: Most papers that employ Differences-in-Differences estimation (DD) use many years of data and focus on serially correlated outcomes but ignore that the resulting standard errors are inconsistent. To illustrate the severity of this issue, we randomly generate placebo laws in state-level data on female wages from the Current Population Survey. For each law, we use OLS to compute the DD estimate of its “effect” as well as the standard error of this estimate. These conventional DD standard errors severely understate the standard deviation of the estimators: we find an “effect” significant at the 5 percent level for up to 45 percent of the placebo interventions. We use Monte Carlo simulations to investigate how well existing methods help solve this problem. Econometric corrections that place a specific parametric form on the time-series process do not perform well. Bootstrap (taking into account the autocorrelation of the data) works well when the number of states is large enough. Two corrections based on asymptotic approximation of the variance-covariance matrix work well for moderate numbers of states and one correction that collapses the time series information into a “pre”- and “post”-period and explicitly takes into account the effective sample size works well even for small numbers of states.
TL;DR: The Robinson Crusoe Fallacy as discussed by the authors is a common error in decision theory, and it stems from inappropriate use of the concept of probability in regulation, international economic sanctions, and organization theory.
Abstract: The decision to stay at home when you have no umbrella and rain is probable is an appropriate problem for decision theory. The decision to speed when you are in a hurry and the police might be patrolling is a game against a rational opponent. Treating the latter like a problem for decision theory is what I call the Robinson Crusoe fallacy. It is quite common and leads to incorrect conclusions. If the game has no pure strategy equilibrium, changes in the payoffs to a player affect not that player's strategy but the strategy of the opponent in equilibrium. For example, modifying the size of the penalty does not affect the frequency of crime commitment at equilibrium, but rather the frequency of law enforcement. I provide examples of this fallacy in regulation, international economic sanctions, and organization theory and argue that it stems from inappropriate use of the concept of probability.