Search

In the past 25 years, computer scientists and statisticians developed machine learning algorithms capable of modeling highly non-linear transformations and interactions of input features. While actuaries use GLMs frequently in practice, only in the past few years have they begun studying these newer algorithms to tackle insurance-related tasks. This work aims to review the applications of machine learning to the actuarial science field and present the current state-of-the-art in ratemaking and reserving. It first gives an overview of machine learning algorithms, then briefly outlines their applications in actuarial science tasks. Finally, the paper summarizes the future trends of machine learning for the insurance industry.

Empirical findings suggest a positive correlation between inequality and social immobility, a phenomenon coined the (Great) Gatsby curve. However, complete explanations of the phenomenon have not yet been proposed. This paper answers two questions: What are Gatsby curves? When do they exist? We build a theoretical environment in which parental investment and education improve the economic prospects of children. Gatsbian economies and Gatsby curves are formally defined, and we characterize the conditions under which they arise. We show that an economy may go from being Gatsbian to non-Gatsbian, and vice versa. JEL Codes: D31, H52, J31, J62

Summary Insulin-like growth factor-binding protein (IGFBP)-2 is a circulating biomarker of cardiometabolic health. Here, we report that circulating IGFBP-2 concentrations robustly increase after different bariatric procedures in humans, reaching higher levels after biliopancreatic diversion with duodenal switch (BPD-DS) than after Roux-en-Y gastric bypass (RYGB) and sleeve gastrectomy (SG). This increase is closely associated with insulin sensitization. In mice and rats, BPD-DS and RYGB operations also increase circulating IGFBP-2 levels, which are not affected by SG or caloric restriction. In mice, Igfbp2 deficiency significantly impairs surgery-induced loss in adiposity and early improvement in insulin sensitivity but does not affect long-term enhancement in glucose homeostasis. This study demonstrates that the modulation of circulating IGFBP-2 may play a role in the early improvement of insulin sensitivity and loss of adiposity brought about by bariatric surgery., Graphical abstract, Highlights IGFBP-2 is modulated by different bariatric surgeries in both human and rodents In humans, IGFBP-2 levels are closely related to insulin sensitization In mice, IGFBP-2 partly mediates surgery-induced weight loss IGFBP-2 is implicated in surgery-induced early improvements in glucose homeostasis, Faramia et al. show that the robust increase in IGFBP-2 levels is a common feature of bariatric procedures in humans and rodents. In patients, this increase closely correlates with insulin sensitization. In mice, IGFBP-2 deficiency partly impairs weight loss and early improvements in glucose homeostasis induced by Roux-en-Y surgery.

The Al-Mg base 5xxx Al alloys are non-heat-treatable. They derive their strength from solid solution and strain hardening. They are, however, supersaturated and this can cause sensitization issues. This work shows that by applying a recently developed cyclic strengthening approach, a high density of Mg-Al clusters can be formed in these materials resulting yield and tensile strengths, and elongations, superior to the precipitation hardened AA6061 (Al-Mg-Si) in the peak aged (T6 state). The implications for categorizations of Al alloys are discussed and future directions of interest are highlighted.

In the mid-nineteenth century, the founder of the American Society for the Prevention of Cruelty to Animals (ASPCA), Henry Bergh, saw criminal punishment as the lynchpin of the protection of animals. Bergh lobbied the New York legislature for the adoption of animal cruelty laws, and took it on himself to enforce those laws. Animal law has evolved considerably since then, but Bergh’s tactics have experienced a renaissance. The animal protection movement’s reliance on criminal law and incarceration to prop up animal status is the subject of a book-length critique by Justin Marceau in Beyond Cages: Animal Law and Criminal Punishment. Picking up on the book’s call for greater scholarly attention to the relationship between criminal justice and animal protection, this essay focuses scrutiny on three aspects of the modern animal protection’s fixation with criminal justice: (1) the animal protection movement’s renewed interest in privatizing the prosecutorial function; (2) the view that by framing the animal as a victim, social change will be more readily possible; and (3) more generally, the view that prosecutors will serve as catalysts for the sort of radical social change the animal protection movement is pursuing.

Through original quantitative research we show that persons convicted of killing a white victim and sentenced to death are more likely to be executed than persons convicted and sentenced to death for killing a black victim. Previous research documents numerous forms of arbitrariness and racial disparity in the administration of the modern death penalty, but focuses exclusively on the charging and sentencing patterns of prosecutors and juries. Previous research also reveals that implicit bias operates within the institutions tasked with seeking and obtaining sentences of death. Our original research shows that the problem of disparate racial outcomes is actually exacerbated through the work of our most trusted check on the death penalty, appellate courts. Building on David Baldus’s storied dataset from Georgia, we demonstrate that the racial disparities he discovered in the penultimate stage of the case – death sentences – were amplified in the ultimate stage of the case – executions. Combining both phases reveals a stunning pattern: the execution rate is roughly 17 times greater in white victim cases than black victim cases. Although Baldus could not have known how the cases would unfold post-sentencing, our findings indicate that the racial disparities described in McCleskey v Kemp (1987) underestimated the extent of the death penalty’s arbitrariness problem.

In actuarial science, collective risk models, in which the aggregate claim amount of a portfolio is defined in terms of random sums, play a crucial role. In these models, it is common to assume that the number of claims and their amounts are independent, even if this might not always be the case. We consider collective risk models with different dependence structures. Due to the importance of such distributions in an actuarial setting, we first investigate a collective risk model with dependence involving the family of multivariate mixed Erlang distributions. Other models based on mixtures involving bivariate and multivariate copulas in a more general setting are then presented. These different structures allow to link the number of claims to each claim amount, and to quantify the aggregate claim loss. Then, we use Archimedean and hierarchical Archimedean copulas in collective risk models, to model the dependence between the claim number random variable and the claim amount random variables involved in the random sum. Such dependence structures allow us to derive a computational methodology for the assessment of the aggregate claim amount. While being very flexible, this methodology is easy to implement, and can easily fit more complicated hierarchical structures.

In this paper, we investigate dependent risk models in which the dependence structure is defined by an Archimedean copula. Using such a structure with specific marginals, we derive explicit expressions for the pdf of the aggregated risk and other related quantities. The common mixture representation of Archimedean copulas is at the basis of a computational strategy proposed to find exact or approximated values of the distribution of the sum of risks in a general setup. Such results are then used to investigate risk models in regard to aggregation, capital allocation and ruin problems. An extension to nested Archimedean copulas is also discussed.

Variable annuity guarantees, and particularly guaranteed lifetime withdrawal benefit (GLWB) guarantees, have become very important in the wealth management industry. These guarantees, which provide to clients revenue protection while allowing them to retain equity market participation, exhibit significant systematic risks from the issuer's standpoint. Risk management of GLWB guarantees thus is a main concern for insurance companies. This paper assesses the impact of the guarantee liability modeling on the hedge efficiency of GLWB guarantees with respect to three significant systematic risks for these guarantees, that is to say, the stock market, interest rate and longevity risks. The present work thus aims to extend the hedge efficiency analysis performed in Kling et al (2011), which focuses on the stock market risk. In this paper, stochastic interest rates are shown to have primary importance in the guarantee liability modeling of GLWB guarantees. This paper also analyzes the impact of the outer loop modeling of mortality on the hedge efficiency of GLWB guarantees. A risk allocation between financial and longevity risks is used to show that longevity holds a significant share of the total risk of a hedged GLWB guarantee.

The pervasiveness of digital video image capture by large segments of the public has produced a wide range of interesting social challenges, but also presents provocative new opportunities for free speech, transparency, and the promotion of democracy. The opportunity to gather and disseminate images, facilitated by the reduced expense and easy access to camera phones and other hand-held recording devices, decentralizes political power in transformative ways. But other uses of this technology represent potentially significant intrusions on property rights and personal privacy. This tension creates a substantial dilemma for policymakers and theorists who care about both free speech and privacy. Because of these putative social interests, laws governing video image capture are becoming more widespread across a number of different regulatory regimes, from Federal Aviation Administration regulations of drone recordings to bans on recordings of police officers performing their duties to so-called Ag-Gag laws, which criminalize the video recording of incidents of animal abuse at commercial agricultural facilities. In this Article, we examine constitutional theory and doctrine as applied to emerging government regulations of video image capture and propose a framework that will promote free speech to the fullest extent possible without presenting unnecessary intrusions into privacy interests.The Article first argues that video recording is a form of expression or, at the very least, is conduct that serves as a necessary precursor of expression such that it counts as speech within the meaning of the First Amendment. We continue with the novel argument that none of the features that make video recording a form of speech apply differently when the recording takes place on private property. Next, we examine under what circumstances video recording is constitutionally protected. We claim that video recording in public places or on private property with the consent of those recorded is presumptively protected speech under the First Amendment. But we also argue that the right to record attaches even when the recording is nonconsensual and occurs on private property, as long as the material recorded is a matter of public concern. While we acknowledge that the First Amendment does not limit the enforceability of generally applicable prohibitions on access to private property, we nonetheless suggest that a recording of activity that is a matter of public concern and is done by someone who is lawfully present on that private property is protected speech.That is not to say that all regulation of such recordings violates the First Amendment, and we therefore address when countervailing governmental interests might justify limitations on the right to record, including tangible property interests and reasonable privacy expectations. Throughout this part, we draw on examples of laws regulating video recordings to suggest how our proposed model for a right to record would apply in context. The First Amendment model developed in this.

Professor Robert J. Smith encourages readers, lawyers, and courts to forget Furman v. Georgia and to focus instead on death penalty challenges grounded in the diminished culpability of nearly all capital defendants. We applaud Professor Smith’s call to focus on the mental and emotional characteristics that reduce the blameworthiness of so many of those charged with capital crimes; recognizing diminished culpability as the rule rather than the exception among capital defendants conveys a reality that rarely finds its way into reported cases. We are troubled, however, by Professor Smith’s call to “forget Furman.” We believe the title and the article’s efforts to undermine Furman-based challenges disserve Professor Smith’s principal goal — addressing the United States’ broken death penalty system.

Lying has a complicated relationship with the First Amendment. It is beyond question that some lies – such as perjury or pretending to be a police officer – are not covered by the First Amendment. But it is equally clear that some lies, even intentionally lying about military honors, are entitled to First Amendment protection. U.S. v. Alvarez, 132 S. Ct. 2537 (2012). To date, however, both Supreme Court doctrine and academic commentary has taken for granted that any constitutional protection for lies is purely prophylactic – it protects the liar to avoid chilling truthful speech. This Article is the first to argue, contrary to conventional wisdom, that certain types of lies paradoxically advance the values underlying the First Amendment. Our framework is descriptively novel and doctrinally important insofar as we provide the first comprehensive post-Alvarez look at the wide range of lies that may raise First Amendment issues. Because there was no majority opinion in Alvarez, there is uncertainty about which standard of constitutional scrutiny should apply to protected lies, an issue we examine at length. Moreover, our normative claim is straightforward: when a lie has intrinsic or instrumental value it should be treated differently from other types of lies and warrant the greatest constitutional protection. Specifically, we argue that investigative deceptions – lies used to secure truthful factual information about matters of public concern – deserve the utmost constitutional protection because they advance the underling purposes of free speech: they enhance political discourse, help reveal the truth, and promote individual autonomy. A prototypical investigative deception is the sort of misrepresentation required in order for an undercover journalist, investigator, or activist to gain access to information or images of great political significance that would not be available if the investigator disclosed her reporting or political objectives. Tactical use of such lies have a long history in American journalism and activism, from Upton Sinclair to his modern day heirs. Using the proliferation of anti-whistleblower statutes like Ag Gag laws as an illustrative example, we argue that investigative deceptions are a category of high value lies that ought to receive rigorous protection under the First Amendment. At the same time, we recognize that not all lies are alike and that in other areas, the government regulation of lies serves legitimate interests. We therefore conclude the Article by drawing some limiting principles to our theory.

In its 1972 Furman v. Georgia decision, the Supreme Court - concerned that the death penalty was being imposed infrequently and without objectively measurable criteria - held that the penalty violated the Eighth Amendment to the Constitution. In the four decades since Furman there has been considerable Eighth Amendment litigation regarding capital punishment, but almost none of it has focused on the Court’s concern with arbitrariness and infrequency. But this may be about to change. With a growing body of quantitative data regarding the low death sentencing rates in several states, Furman is poised to return to center stage. While previous challenges attacked the form of various state capital statutes, new empirical data is leading condemned inmates to challenge the application of state sentencing statutes. This article announces the return of Furman - a splintered opinion that nonetheless remains binding precedent 42 years after it was decided - and provides a reading of that case that can guide courts as they consider the latest round of challenges to the application of capital punishment. A careful revisiting of Furman is necessary and overdue because the critical underpinnings of American death penalty jurisprudence - narrowing, eligibility, and individualization - are currently being conflated, or forgotten altogether by both courts and capital litigants. This Article, is a timely guidepost for the inevitable next wave of Furman litigation.

Legal fields as divergent as family law, torts, contracts, and trusts have each, to varying degrees, addressed the unique legal status of pets. The rights and obligations of pet owners are a topic of increasing legal interest. Even the criminal law has grappled with the uniqueness of animals to a limited extent by criminalizing animal abuse. Legal developments such as these tend to ameliorate the anachronistic view that animals are merely property. However, substantial pockets of the law have not yet grappled with the unique status of animals as something more than property but, perhaps, less than human. This Article is the first to analyze the operation of the criminal defenses — the doctrines of exculpation — for persons who use serious, and even lethal, force in defense of their pets. By exploring the intersection of criminal defenses and the status of animals, there is much to be learned about the ambiguities in our common law doctrines of exculpation and the status of animals in America. The Article is less an argument for greater animal rights (or increased violence) and more a call to understand how the law’s current treatment of pets and pet owners is discordant with our social values and in need of reassessment.

Beginning in 1977, the U.S. Supreme Court instructed lawyers and lower courts that when there is no majority decision “in support of the judgment . . . , the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds.” For decades, commentators and judges alike have vocally lamented the opaque and seemingly intractable nature of this instruction, known as the Marks rule. The usual academic trope in this field consists of a discussion of a recent plurality decision, followed by an account of how difficult it is to discern the narrowest grounds for that decision, and concluding with a statement about how the lack of clarity as to the relevant precedent impedes the Court’s lawmaking function and diminishes the Court’s credibility with the public. By contrast, this Article provides a new framework for understanding plurality precedent. Rather than emphasizing the problems presented by uncertain precedent under the narrowest grounds test, this Article highlights the effectiveness of the rule in simultaneously maintaining judicial credibility with the public while facilitating flexibility for lower court judges. That is to say, this Article celebrates the Marks rule’s success in establishing an acoustic separation between the rule as it is transmitted to lower court judges, and the rule as it is understood by the public. The external message reassures society that there has been no breakdown in the judicial process when a plurality occurs, and the internal, decisional rule is that in the absence of lower court consensus, there is no plurality precedent. In short, Marks effectively reconciles the competing interests of public legitimacy and legal flexibility.

Federal habeas review of criminal convictions is not supposed to be a second opportunity to adjudge guilt. It has been said, by Oliver Wendall Holmes among others, that the sole question on federal habeas is whether the prisoner’s constitutional rights were violated. By the early 1970s, however, scholars criticized this rights-based view of habeas and sounded the alarm that post-conviction review had become too far removed from questions of innocence. Most famously, in 1970 Judge Friendly criticized the breadth of habeas corpus by posing a single question: Is innocence irrelevant? In his view habeas review that focused exclusively on questions of rights in isolation from questions of innocence was misguided.Over the last forty years the habeas landscape has changed so dramatically — both through statutory and common law limits on the writ — that it is appropriate to ask a very different question: Is guilt dispositive? Both substantive law and habeas procedure has evolved so as to substantially disadvantage a guilty habeas petitioner. In many cases regardless of the merits of the constitutional claim, strong evidence of guilt is dispositive in ensuring that relief is denied. A recent trilogy of cases from the last couple of years — Holland v. Florida, Maples v. Thomas, and most importantly, Martinez v. Ryan — signal a potential shift in the Court’s innocence orientation. This Article explores the potential impact of these decisions and in particular argues that they may provide a roadmap for a proceduralist approach to modern habeas — that is, habeas review that prioritizes fair procedures over innocence. The impact of Friendly’s call for greater focus on innocence was gradual but profound, and this Article argues that the Martinez-trilogy may be similarly important in reversing habeas’ four-decade-long infatuation with innocence.

This article reports the conclusions of an empirical study of every murder conviction in Colorado between January 1, 1999 and December 31, 2010. Our goal was to determine: 1) What percentage of first degree murderers in Colorado were eligible for the death penalty; and 2) How often the death penalty was sought against these killers. More importantly, our broader purpose was to determine whether Colorado’s statutory aggravating factors meaningfully narrow the class of death eligible offenders as required by the Constitution. We discovered that while the death penalty was an option in approximately ninety two percent of all first degree murders, it was sought by the prosecution initially in only three percent of those killings, pursued all the way through sentencing in only one percent of those killings, and obtained in only 0.6 percent of all cases. These numbers compel the conclusion that Colorado’s capital sentencing system fails to satisfy the constitutional imperative of creating clear, statutory standards for distinguishing between the few who are executed and the many who commit murder. The Eighth Amendment requires that these determinations of life and death be made at the level of reasoned legislative judgment, and not on an ad hoc basis by prosecutors. The Supreme Court has emphasized that a State’s capital sentencing statute must serve the “constitutionally necessary function . . . [of] circumscrib[ing] the class of persons eligible for the death penalty” such that only the very worst killers are eligible for the law’s ultimate punishment. Colorado’s system is unconstitutional under this standard because nearly all first degree murderers are statutorily eligible to be executed.

The near-unanimous belief among courts and commentators that the criminal law contains a voluntary act requirement obscures deep disagreement about the meaning, scope, and application of that requirement. This Article explores these longstanding and vexing theoretical disagreements, identifies practical problems that result from these uncertainties, and suggests a novel framework for substantially resolving the confusion. The resolution of these questions is no mere academic exercise. The difference in many cases, particularly strict liability prosecutions, will be the difference between guilt and innocence.

The recent decisions in Missouri v. Frye and Lafler v. Cooper represent a seismic shift in the Court’s right to counsel jurisprudence. No longer is the right to counsel limited to protecting thefairness and adequacy of the trial. Although these two cases arose in the plea bargaining context, the doctrinal shift may have its greatest impact in cases where plea bargaining is not at issue.This Article identifies the salient features of this new — non-trial oriented — conception of the right to counsel and explains its far-reaching impacts on the day-to-day practice of criminal law.Specifically, this Article explains the import of the newly minted right to effective assistance as it relates to a variety of procedural constitutional rights, including speedy trial, pretrial detention, double jeopardy, and jury selection rights. The explicit recognition that the right to counsel is not only, or even primarily a trial or truth protecting right promises to be a staggeringly important constitutional event.

Current developments in international law witness an increase number of disputes arising out of conflicting societal values, such as the promotion of economic development versus the protection of health and the environment. Irrespective of its precise normative content, sustainable development, characterized as an "interstitial norm" or a "meta-principle", can assist in resolving these disputes in a more harmonious fashion. This chapter seeks to explore the role that sustainable development has played in the rules of the WTO. We argue that the objective of sustainable development has reshaped the interpretation of key provisions within the WTO Agreement, offering to the Members more policy space to protect essential interests. We contend that the concept has operated at the interstices of two competing priorities in WTO law: the right to free trade and economic development and the right to restrict trade in order to protect life, health, and environment.

When it was decided in 2002, Ring v. Arizona appeared to be a watershed in the way capital sentences are handed out in the United States: it overturned several states’ death penalty statutes and appeared to imperil many more. Ring announced that the rule of Apprendi v. New Jersey applied to capital sentencing and required that any fact necessary to the imposition of the death penalty be proven to a jury and beyond a reasonable doubt. Yet eight years after the case was decided, it is not clear what, if anything, Ring in fact demands of the states. Determining exactly what constitutes fact-finding, and therefore which tasks must be carried out by the capital jury rather than a judge, remains a challenging task. In this article we investigate the impact of Ring by analyzing several typical capital statutes against both the language of the Ring opinion and the broader framework of the Court’s Sixth and Eighth Amendment jurisprudence. What we find is that in all but the most obvious cases, Ring’s mandate is an extraordinarily weak one. What is more, Ring creates perverse incentives for the states; juries can be removed from the capital sentencing equation entirely simply by making capital decisionmaking open-ended and amorphous rather than fact-based and legalistic. We argue that this pressure toward open-ended decisionmaking creates significant Eighth Amendment concerns. We conclude that this tension between the Sixth and Eighth Amendments demonstrates the limits of Ring’s narrow focus on fact-finding and argues for the grounding of a more robust capital jury right in the Eighth Amendment.

Countless articles and judicial opinions have been devoted to the task of deciphering the scope and application of the limitations on habeas corpus relief announced in the Anti-Terrorism and Effective Death Penalty Act of 1996 (AEDPA). Over the past ten years courts and scholars have developed an intricate framework of analysis for nearly every sub-section of § 2254. The decade-long process of interpretation and commentary has been characterized by questions of statutory meaning and federalism that appear to be as irresolvable for courts as they are intriguing to academics. But in the rush to sort out the minutia of AEDPA, the hallmarks of our legal system, basic due process and constitutional supremacy, have been overlooked. This Article aims to re-focus the debate. The application and discussion of AEDPA’s limitations on relief has devolved into a bitter argument over the meaning of a statute which lacks a discoverable meaning, much less an obvious or plain meaning. It is statutory esotericism or statutory obfuscation much more than it is statutory interpretation. The discussion has become so technical and specialized, not to mention politically polarized, that we are at risk of permanently overshadowing the historical and constitutional underpinnings of the Great Writ. The goal of this Article is to recast and simplify the habeas debate and achieve some much needed common ground. The thesis is simple: where the state post-conviction process does not provide a meaningful corrective process such that federal constitutional issues are not “fully and fairly” adjudicated, it is necessary for the federal courts to review the federal claims de novo. This modest procedural proposal is compelled by due process through a celebrated line of cases, and yet in the frenzy to interpret § 2254 – in working out all of the (e)(2)’s and the (d)(1)’s – we have forgotten due process. It is time to return to it. More than a century ago in Frank v. Magnum and Moore v. Dempsey, the Court recognized the critical role that federal habeas review must play in ensuring that basic constitutional criminal procedure rights were adjudicated in a minimally “full and fair” manner by state courts. To be sure this fair-process check on state adjudications was of minimal, even trivial, significance during the Brown v. Allen era when federal habeas was viewed by the Court as providing a virtually unchecked opportunity to rework the underlying merits of the state adjudication. But the limitations on federal habeas review born during the Rehnquist and Burger Courts and enhanced through the enactment of the AEDPA compel a reasoned revisiting of due process in this context. After surveying the law defining the minimum standards of due process in the context of adjudicating constitutional criminal procedure rights – the “full and fair” mandate – this paper recommends a reading of §2254 that is both faithful to due process and consistent with the goals of the AEDPA.

We develop a model of a competitive rental housing market with an endogenous rate of tenancy default arising from income uncertainty. Potential tenants must choose to engage in a costly search for rental housing, and must commit to a rental agreement before the uncertainty is resolved. We show that there are two possible equilibria in this market: a market-clearing equilibrium and an equilibrium with excess demand. Therefore, individuals might not have access to rental housing because they are unable to afford to look for housing, they are unable to pay their rent, or with excess demand in the market they are simply unable to find a rental unit. We show that government regulations affecting the cost of default to the housing suppliers and the quality of rental units can have different effects on the equilibrium variables of interest - rental rate, quantity demanded and supplied, and access to rental housing - depending on the type of equilibria in the market. A numerical example illustrates these results.

We consider the implications of ethical behaviour on the effect of a redistributive tax-transfer system. In choosing their labour supplies, individuals take into account whether their tax liabilities correspond to what they view as ethically acceptable. If tax liabilities are viewed as ethically acceptable, a taxpayer behaves ethically, does not distort her behaviour, and chooses to work as if she were not taxed. On the other hand, if ethical behaviour results in tax liabilities that exceed those that are ethically acceptable, she behaves egoistically (partially or fully), distorts her behaviour, and chooses her labour supply taking into account the income tax. We establish taxpayers' equilibrium behaviour and obtain that labour supply is less elastic when taxpayers may behave ethically than when they act egoistically. We characterize and compare the egoistic voting equilibrium linear tax schedules under potentially ethical and egoistic behaviour. We also compare our results to those obtained under altruism, an alternative benchmark.

We consider a world in which the mode of food production, foraging or agriculture, is endogenous, and in which technology grows exogenously. Within a model of coalition formation, we allow individuals to form co-operative communities (bands) of foragers or farmers rationally. At the lowest levels of technology, equilibrium entails the grand coalition of foragers, a co-operative structure which avoids over-exploitation of the environment. But at a critical state of technology, the co-operative structure breaks down through an individually rational splintering of the band. At this stage, there can be an increase in work and through the over-exploitation of the environment, a food crisis. In the end, technological growth may lead to a one-way transition from foraging to agriculture.

This paper studies inter-jurisdictional competition in the fight against crime and its impact on occupational choice and the allocation of capital. In a world where capital is mobile, jurisdictions are inhabited by individuals who choose to become either workers or criminals. Because the return of the two occupations depends on capital, and because investment in capital in a jurisdiction depends on its crime rate, there is a bi-directional relationship between capital investment and crime which may lead to capital concentration. By investing in costly law enforcement, a jurisdiction makes the choice of becoming a criminal less attractive, which in turn reduces the number of criminals and makes its territory more secure. This increased security increases the attractiveness of the jurisdiction for investors and this can eventually translate into more capital being invested. We characterize the Nash equilibria—some entailing a symmetric outcome, others an asymmetric one—and study their efficiency.

A two-region economy consists of a given but different number of immobile workers in each region, and a given number of mobile firms. Firms create jobs where they locate, but there is frictional unemployment. Two sorts of agglomeration effects arise: those from economies of scale in matching, and those from production economies external to the firm. Regions may either be part of a unitary state in which case all regional policies are decided by the central government, or they may be part of a federal state in which case some policies are determined by the regional governments. We characterize the resource allocations in both a unitary and a federal state, and identify the set of instruments that are required to replicate the social optimum in each state.