Search the archive.

The democratic peace — the idea that democracies rarely fight one another — has been called “the closest thing we have to an empirical law in the study of international relations.” Yet, some contend that this relationship is spurious and suggest alternative explanations. Unfortunately, in the absence of randomized experiments, we can never rule out the possible existence of such confounding biases. Rather than commonly used regression-based approaches, we apply a nonparametric sensitivity analysis. We show that overturning the positive association between democracy and peace would require a confounder that is 47 times more prevalent in democratic dyads than in other dyads. To put this number in context, the relationship between democracy and peace is at least five times as robust as that between smoking and lung cancer. To explain away the democratic peace, therefore, scholars must find far more powerful confounders than already those identified in the literature.

The risk of war by miscalculation is a popular scholarly subject and a frequent concern of policymakers. Theories of war by miscalculation assume either ignorance by decisionmakers, bad faith by policy participants (usually the military), or a mechanistic series of exchanges that obviate the need for policymakers to display judgment and responsibility. Yet historians struggle to produce evidence that political leaders were either ignorant of the risks they were choosing or emasculated from any ability to affect the course of events. Theories of accidental war are, as Geoffrey Blainey has said, “a description masquerading as an explanation.”[1] Instead of explaining historical events, the myth of accidental war provides an intricate puzzle for intellectuals, a justification for moralists to fear military subversion of political decisionmakers, and an evasion of culpability for those decisionmakers.

The debate is not esoteric—it has urgent policy applications. U.S. policy regarding Russia’s February 2022 full-scale invasion of Ukraine has been heavily influenced by U.S. President Joe Biden and his administration’s concerns about “a third world war.”[2] Russia violated the sovereignty and the internationally recognized territory of Ukraine, despite its commitment to respect Ukraine’s security in the 1995 Budapest Memorandum. Failing to achieve its political objectives on the battlefield, Russian strategy shifted to targeting civilian dwellings, destroying infrastructure, and committing mass atrocities. All of these actions violate Russia’s international commitments and overtly challenge the Biden administration’s leadership of the rules-based international order, as well as its pledge to “stand against human rights abuses wherever they occur.”[3]

Today, when we talk about global order, we often lament the divisions inside the United Nations Security Council. This year marks the 50th anniversary of important unity: when the U.N. Security Council unanimously recommended German membership in the United Nations. The German Democratic Republic and the Federal Republic of Germany both entered the U.N. in September 1973.

Germany today is the U.N.’s fourth-largest financial contributor and has been elected to six different Security Council rotations. Germany, along with Namibia, is currently facilitating consultations for the U.N.’s 2024 “Summit of the Future,” to consider how the multilateral system should address old and new challenges.

Playing with Wild Cards Without doubt, simulations -- including nonstandard ones -- can aid planning. The question is: To what end? And to what effect? Exploring "wild cards" in order to identify warning signs or to define limits is one thing; using them to establish force structure or modernization requirements, quite another. Especially suspect would be using scenarios that are detached from declared US interests to define current requirements; this would put the military "cart" before the political "horse." Another, broader concern is how the effusion of improbable conflict scenarios affects public policy discourse overall. Conflict scenarios, both wild and tame, can gain more credibility in the telling than they deserve. Cognitive researcher Massimo Piattelli-Palmarini calls this the "Othello effect," referring to the trail of plausible but false suppositions that led Othello to murder his wife, Desdemona. Even the most farfetched scenarios comprise a number of steps or links each of which may seem plausible or even probable given the one that came before. Although the likelihood of the scenario dwindles with each step, the residual impression is one of plausibility. Omitted are the many branches at each step that would lead to a neutral or even positive outcome. The resulting snapshots, although numerous, offer a highly-selective view of what the future may hold. And the fact that only the negative outcomes are articulated and exercised can distort the general public impression of risk. Living with Uncertainty There is no escape from uncertainty, but there is relief from uncertainty hysteria. It begins with recognizing that instability has boundaries -- just as turbulence in physical systems has discernable onset points and parameters. The turbulence of a river, for instance, corresponds to flow and to the contours of the river's bed and banks. It occurs in patches and not randomly. The weather also is a chaotic system that resists precise long-range forecasting, but allows useful prediction of broader trends and limits. Despite uncertainty, statements of probability matter. They indicate the weight of evidence -- or whether there is any evidence at all. The uncertainty hawks would flood our concern with a horde of dangers that pass their permissive test of "non-zero probability." However, by lowering the threshold of alarm, they establish an impossible standard of defense sufficiency: absolute and certain military security. Given finite resources and competing ends, something less will have to do. Strategic wisdom begins with the setting of priorities -- and priorities demand strict attention to what appears likely and what does not.

Third is no great power conflict –

The problem of the second method is that it is very difficult to "calculate" politically unacceptable losses. If the risk of terrorism is defined in traditional terms by probability and potential loss, then the focus on dramatic terror attacks leads to the marginalization of probabilities. The reason is that even the highest degree of improbability becomes irrelevant as the measure of loss goes to infinity. ^o The mathematical calculation of the risk of terrorism thus tends to overestimate and to dramatize the danger. This has consequences beyond the actual risk assessment for the formulation and execution of "risk policies": If one factor of the risk calculation approaches infinity (e.g., if a case of nuclear terrorism is envisaged), then there is no balanced measure for antiterrorist efforts, and risk management as a rational endeavor breaks down. Under the historical condition of bipolarity, the "ultimate" threat with nuclear weapons could be balanced by a similar counterthreat, and new equilibria could be achieved, albeit on higher levels of nuclear overkill. Under the new condition of uncertainty, no such rational balancing is possible since knowledge about actors, their motives and capabilities, is largely absent. The second form of security policy that emerges when the deterrence model collapses mirrors the "social probability" approach. It represents a logic of catastrophe. In contrast to risk management framed in line with logical probability theory, the logic of catastrophe does not attempt to provide means of absorbing uncertainty. Rather, it takes uncertainty as constitutive for the logic itself; uncertainty is a crucial precondition for catastrophes. In particular, catastrophes happen at once, without a warning, but with major implications for the world polity. In this category, we find the impact of meteorites. Mars attacks, the tsunami in South East Asia, and 9/11. To conceive of terrorism as catastrophe has consequences for the formulation of an adequate security policy. Since catastrophes hap-pen irrespectively of human activity or inactivity, no political action could possibly prevent them. Of course, there are precautions that can be taken, but the framing of terrorist attack as a catastrophe points to spatial and temporal characteristics that are beyond "rationality." Thus, political decision makers are exempted from the responsibility to provide security—as long as they at least try to preempt an attack. Interestingly enough, 9/11 was framed as catastrophe in various commissions dealing with the question of who was responsible and whether it could have been prevented. This makes clear that under the condition of uncertainty, there are no objective criteria that could serve as an anchor for measuring dangers and assessing the quality of political responses. For ex- ample, as much as one might object to certain measures by the US administration, it is almost impossible to "measure" the success of countermeasures. Of course, there might be a subjective assessment of specific shortcomings or failures, but there is no "common" currency to evaluate them. As a consequence, the framework of the security dilemma fails to capture the basic uncertainties. Pushing the door open for the security paradox, the main problem of security analysis then becomes the question how to integrate dangers in risk assessments and security policies about which simply nothing is known. In the mid 1990s, a Rand study entitled "New Challenges for Defense Planning" addressed this issue arguing that "most striking is the fact that we do not even know who or what will constitute the most serious future threat, "^i In order to cope with this challenge it would be essential, another Rand researcher wrote, to break free from the "tyranny" of plausible scenario planning. The decisive step would be to create "discontinuous scenarios ... in which there is no plausible audit trail or storyline from current events"52 These nonstandard scenarios were later called "wild cards" and became important in the current US strategic discourse. They justified the transformation from a threat-based toward a capability- based defense planning strategy.53 The problem with this kind of risk assessment is, however, that even the most absurd scenarios can gain plausibility. By constructing a chain of potentialities, improbable events are linked and brought into the realm of the possible, if not even the probable. "Although the likelihood of the scenario dwindles with each step, the residual impression is one of plausibility. "54 This so-called Othello effect has been effective in the dawn of the recent war in Iraq. The connection between Saddam Hussein and Al Qaeda that the US government tried to prove was disputed from the very beginning. False evidence was again and again presented and refuted, but this did not prevent the administration from presenting as the main rationale for war the improbable yet possible connection between Iraq and the terrorist network and the improbable yet possible proliferation of an improbable yet possible nuclear weapon into the hands of Bin Laden. As Donald Rumsfeld famously said: "Absence of evidence is not evidence of absence." This sentence indicates that under the condition of genuine uncertainty, different evidence criteria prevail than in situations where security problems can be assessed with relative certainty.

It may have once been the case that being attacked by another country was a major threat to the lives of ordinary people. It may also be true that there are still some pretty serious dangers out there associated with the spread of nuclear weapons. For the most part, however, most of what you've been told about national security and all the big threats which can supposedly kill you is one big con designed to distract you from the things that can really hurt you, such as the poverty, inequality and structural violence of capitalism, global warming, and the manufacture and proliferation of weapons - among others. The facts are simple and irrefutable: you're far more likely to die from lack of health care provision than you are from terrorism; from stress and overwork than Iranian or North Korean nuclear missiles; from lack of road safety than from illegal immigrants; from mental illness and suicide than from computer hackers; from domestic violence than from asylum seekers; from the misuse of legal medicines and alcohol abuse than from international drug lords. And yet, politicians and the servile media spend most of their time talking about the threats posed by terrorism, immigration, asylum seekers, the international drug trade, the nuclear programmes of Iran and North Korea, computer hackers, animal rights activism, the threat of China, and a host of other issues which are all about as equally unlikely to affect the health and well-being of you and your family. Along with this obsessive and perennial discussion of so-called 'national security issues', the state spends truly vast sums on security measures which have virtually no impact on the actual risk of dying from these threats, and then engages in massive displays of 'security theatre' designed to show just how seriously the state takes these threats - such as the x-ray machines and security measures in every public building, surveillance cameras everywhere, missile launchers in urban areas, drones in Afghanistan, armed police in airports, and a thousand other things. This display is meant to convince you that these threats are really, really serious. And while all this is going on, the rulers of society are hoping that you won't notice that increasing social and economic inequality in society leads to increased ill health for a growing underclass; that suicide and crime always rise when unemployment rises; that workplaces remain highly dangerous and kill and maim hundreds of people per year; that there are preventable diseases which plague the poorer sections of society; that domestic violence kills and injures thousands of women and children annually; and that globally, poverty and preventable disease kills tens of millions of people needlessly every year. In other words, they are hoping that you won't notice how much structural violence there is in the world. More than this, they are hoping that you won't notice that while literally trillions of dollars are spent on military weapons, foreign wars and security theatre (which also arguably do nothing to make any us any safer, and may even make us marginally less safe), that domestic violence programmes struggle to provide even minimal support for women and children at risk of serious harm from their partners; that underfunded mental health programmes mean long waiting lists to receive basic care for at-risk individuals; that drug and alcohol rehabilitation programmes lack the funding to match the demand for help; that welfare measures aimed at reducing inequality have been inadequate for decades; that health and safety measures at many workplaces remain insufficiently resourced; and that measures to tackle global warming and developing alternative energy remain hopelessly inadequate. Of course, none of this is surprising. Politicians are a part of the system; they don't want to change it. For them, all the insecurity, death and ill-health caused by capitalist inequality are a price worth paying to keep the basic social structures as they are. A more egalitarian society based on equality, solidarity, and other non-materialist values would not suit their interests, or the special interests of the lobby groups they are indebted to. It is also true that dealing with economic and social inequality, improving public health, changing international structures of inequality, restructuring the military-industrial complex, and making the necessary economic and political changes to deal with global warming will be extremely difficult and will require long-term commitment and determination. For politicians looking towards the next election, it is clearly much easier to paint immigrants as a threat to social order or pontificate about the ongoing danger of terrorists. It is also more exciting for the media than stories about how poor people and people of colour are discriminated against and suffer worse health as a consequence. Viewed from this vantage point, national security is one massive confidence trick - misdirection on an epic scale. Its primary function is to distract you from the structures and inequalities in society which are the real threat to the health and wellbeing of you and your family, and to convince you to be permanently afraid so that you will acquiesce to all the security measures which keep you under state control and keep the military-industrial complex ticking along. Keep this in mind next time you hear a politician talking about the threat of uncontrolled immigration, the risk posed by asylum seekers or the threat of Iran, or the need to expand counter-terrorism powers. The question is: when politicians are talking about national security, what is that they don't want you to think and talk about? What exactly is the misdirection they are engaged in? The truth is, if you think that terrorists or immigrants or asylum seekers or Iran are a greater threat to your safety than the capitalist system, you have been well and truly conned, my friend. Don't believe the hype: you're much more likely to die from any one of several forms of structural violence in society than you are from immigrants or terrorism. Somehow, we need to challenge the politicians on this fact.

Nanomaterials research must overcome the issue of scalability if it is to make a tangible impact in the real world. One of the tenets of nanoscience is the fact that the physical and chemical properties of matter change with size, and as you reach the nanometre length scale new behaviours start to appear. This is largely the appeal of doing science at this scale. We can now study, manipulate and control nano-objects — even one at a time in many cases — to produce a desired effect or a property that cannot be achieved in the bulk. The rush to the nanoscale is almost ubiquitous in science, because nanoscience offers the unique ability to understand physical and chemical processes at the most fundamental level. In principle, this knowledge should be instrumental to improve and optimize a system under investigation. But despite these fundamental advances, nanotechnology is confronted with a critical bottleneck. We are still struggling to translate the fundamental advances reported in the scientific literature into tangible technological applications that can be appreciated at the layman's level. The problem is twofold. First, the properties of matter change when scaled up, just like they change when scaled down to the nanoscale; in particular, the level of control that can be exerted at the nanoscale or at the single-object level tends to wane at the meso- and macroscales or when dealing with a large number of objects. And second, industry is reluctant to invest money in developing new large-scale processes for nanomaterial fabrication unless they are guaranteed a sizeable profitable return. The situation is especially relevant in applied fields, because there is a disconnect between laboratory- and industrial-scale investigation.

Many discussions of U.S. regulations seem to presume that American laws can restrict or prevent AI development. However, this is demonstrably not the case. While the U.S. has led the world in the development of key computing technologies and several of the world’s largest software companies[30]—Microsoft, Google, Oracle, IBM, Apple and Adobe—are American firms, the U.S. is not the only place where AI is being developed. Russian president Vladimir Putin has heralded AI as “the future, not only for Russia, but for all humankind.”[31] In September 2017, he went as far as to tell Russian students that the nation that “becomes the leader in this sphere will become the ruler of the world.”[32]

With Russia and other nations embracing AI,[33] nations that don’t innovate in AI technologies—or worse, those that actually restrict its development—run the risk of falling behind and not being able to compete with the countries that promote AI development.[34] Advanced AIs can create advantages for a nation’s businesses and its defense. Nations without AI or with less mature AI systems might be placed at a severe disadvantage and forced to buy systems with whatever capabilities the more advanced nations are willing to let their firms sell to other countries. While the state of nations after the introduction of AI is inherently unclear, one thing is apparent: restricting AI development in the U.S. won’t stop it from being developed. In fact, this may make it far more likely that the eventual winning AI systems won’t respect our societal values, because they have been developed by another country or group that doesn’t share them.

Regarding the first question, a number of studies have examined how so- called youth bulges contribute to violence and conflict within states, revealing that having a high ratio of people aged 15-24 among the adult population increases the likelihood of domestic terrorism and civil war.3 Yet no study has thus far examined, either theoretically or empirically, how youth bulges alter the probability of military conflict between states. Ours is the first to do so. Our empirical analysis finds that countries with a large number of young people as a proportion of the total population are more likely to engage in interstate conflict than are countries with fewer young people.

As for the second question, some scholars have theorized that old societies are likely to be more peaceful.4 Yet, the theoretical reasons why we would ex¬pect societal aging to lower the likelihood of conflict have not been fully delineated, and no scholar has evaluated this proposition empirically. Our empirical analysis shows that old countries are the least likely to engage in international conflict.

Notwithstanding the moderately positive conclusions on the potential of environmental taxation for enhancing transitions, we identify six barriers and limitations that should not be overlooked. First, public and political support for environmental taxation is limited. People dislike taxes in general (Avi-Yonah & Uhlmann, 2009), and environmental taxation in particular because they see it as an illegitimate sources of revenue for the government (Green Fiscal Commission, 2009a). One way of making environmental taxation and green tax reform more acceptable is using insights from behavioural economics. For example, to make people accept a budget-neutral tax reform, the value of what they receive should be higher than the value of what they sacrifice. This phenomenon is called the endowment effect (Kahneman et al., 1991). Second, an environmental tax in a socio-technical system such as the energy system will need to be implemented for a long period (more than ten years) before it starts living up to its potential, for both technology and practices. However, politically this is a long period, and the environmental tax reforms that survive the first political decision process, are often reversed when a new government takes office. This happened to the Australian carbon tax in 2013 (Carl & Fedor, 2016). A rare counterexample is the Swedish carbon tax, which has been in place since 1991 (Daugbjerg & Svendsen, 2003). Third, choices made for or against a certain technology or practice will influence policy options in the future (path dependencies) (Rotmans, 2003). Misalignment of a tax with the transition vision could result in a lock-in or even system breakdown (van der Brugge & Rotmans, 2007; Kemp & Pontoglio, 2011), although a subsidy poses a larger risk of a lock-in than a tax. Fourth, the lack of sufficient, appropriate or promising niches is a barrier, because niches that do not perform consistently will not provide any benefits (Sopha et al., 2013). Fifth, environmental taxes could have some undesired side effects, such as a negative competitiveness impact and a regressive impact. However, those deficiencies can be largely redressed by careful design and compensating measures as part of a green tax reform (Mikael Skou Andersen & Ekins, 2009; Klenert & Mattauch, 2016). Finally, sixth, a stand-alone policy instrument, taxation or other, will always have a number of downsides, part of which can be overcome by using smart instrument and policy mixes (Oikonomou et al., 2014), such as an environmental tax in combination with a large communication campaign explaining the benefits of the tax to citizens, and with revenue recycling, for example a subsidy, to increase public support.

The decision in the Loper case overturned a 40-year-old precedent that permitted agencies to resolve statutory ambiguities without excessive judicial oversight as long as the agency interpretation was reasonable. That deference is now gone. It should be noted that reliance on the Chevron doctrine had declined, and the court, in many ways, merely confirmed the emerging law. The difficulty with the decision is that it introduces rules that will likely produce instability rather than the clarity the majority claimed to be providing. While agency interpretations will be considered with appropriate respect, when courts review agency actions, agency determinations will no longer be given the deference they had enjoyed.

By overturning the Chevron principle of reasonable (and limited!) deference to federal agencies in rulemaking, the Supreme Court has knocked out a core pillar of American administrative law and created a real risk of regulatory chaos across a range of policy domains, including environmental protection, health care, labor standards, land use, and corporate reporting. In eliminated a long-established policymaking framework and providing almost no guidance for courts to apply in determining whether agencies have acted reasonably in implementing the legislation adopted by the Congress and signed into law by the president, the court may well have unleashed an unprecedented level of uncertainty into the regulatory arena. The chaos that will result as lower courts pass judgment individually on regulations in areas that are often quite technical and in which the judges involved often lack any expertise will translate into risk and cost for all regulated parties and impose, in effect, an uncertainty tax on businesses large and small, which will spillover onto every American.

The Biden administration on Wednesday blamed a staffing shortage at the Internal Revenue Service for enabling the wealthiest 1% of Americans to shirk $163 billion in tax bills annually, pointing to the finding to support its plan for dramatically increasing resources for the tax agency.

More than one-quarter of unpaid taxes comes from just the top 1% of earners in the United States, according to a new report from the Treasury Department, with more than 20% stemming from the top 0.5%. The IRS is picking its battles, Treasury said, creating a separate tax system for low-income Americans and the wealthy who can afford to hire professionals to help them take advantage of IRS' shortcomings.

A $1.66 trillion top-line spending agreement announced by congressional leaders this week will accelerate funding cuts to the Internal Revenue Service favored by Republicans.

The bipartisan deal would cut $10 billion from the IRS during fiscal 2024, one year earlier than Democrats and Republicans previously agreed on in a deal to raise the debt limit last summer. The debt limit deal included $20 billion in overall reductions to a controversial IRS funding boost.

MIT economist Jonathan Gruber, often described as an architect Obamacare, recently said that some of these problems can be alleviated by increasing the “tax” on those without insurance. “I think probably the most important thing experts would agree is we need a larger mandate penalty,” said Gruber.

Depending on how high the penalty goes, there could be a constitutional problem with that. In the opinion that converted the “penalty” into a constitutional “tax,” Chief Justice Roberts described the characteristics of the “shared responsibility payment” that made it, constitutionally speaking, a tax rather than a penalty. One of those characteristics is that the penalty was not too high: “for most Americans the amount due will be far less than the price of insurance, and, by statute, it can never be more. It may often be a reasonable financial decision to make the payment rather than purchase insurance, unlike the ‘prohibitory’ financial punishment in Drexel Furniture.” In Drexel Furniture, also known as the Child Labor Tax Case, the Court struck down a 10 percent tax on the profits of employers who used child labor in certain businesses. One reason the Court struck it down was because its “prohibitory and regulatory effect and purpose are palpable.”

We close with one final potential explanation. It is conceivable that regulators are simply highly risk averse—they understand that they have the authority to employ Pigovian taxes, but they have little incentive to experiment and prefer to opt for tried-and-true methods that have been repeatedly validated. Risk aversion might be coupled with misaligned institutional incentives: it is plausible that an agency head has more to lose from attempting a new type of regulation and having it be rejected (judicially or politically) than she has to gain if the regulation succeeds. The safer path of command-and-control regulation might also be the more personally advantageous one. The uncertainty embedded in agencies’ organic statutes might be heightening these tendencies.

The rarity of Pigovian taxes could also be based in part upon the negative expressive and symbolic nature of such measures. Taxes have long had a negative connotation in American politics, even when they are not being imposed directly upon individuals.162 When the Obama administration began to pursue legislation to reduce greenhouse gas emissions, it rejected Pigovian taxes as politically unpalatable, precisely (and entirely) because of the word “tax” in the name.163 President Clinton encountered stiff resistance to a much more modest energy tax several years earlier.164 This was despite the fact that, as economists have shown, a properly structured Pigovian tax and a properly structured cap-and-trade system will function almost identically.165 Then, when Obama eventually proposed a cap-and-trade system, opponents attacked it as “cap and tax,”166 seeking to tar it with the same brush. The word “tax” is toxic, particularly for political conservatives.167

At the same time, Pigovian taxes have a bad reputation among many political liberals, particularly when compared with the alternative of direct regulation. Part of the opposition relates to the fact that Pigovian taxes allow a company to continue polluting so long as it is willing to pay the appropriate price. If pollution is a moral wrong, in that the polluter is imposing harm upon an unconnected victim, then it strikes some commentators as immoral to permit a firm to continue to pollute on the condition that it pays a fee.168 This objection evokes concerns of compensatory or distributive justice: if a polluter pays a Pigovian tax in order to harm some victim, the tax payment may go to the general fisc rather than the victim herself. Of course, this concern ignores the fact that an ideally designed Pigovian tax would reduce pollution and harm to the same level as an ideal system of command-and-control regulation.

The incentive to lobby for new legislation or regulation dissipates when the regulation takes the form of a Pigovian tax. Of course, a Pigovian tax makes pollution-controlling equipment or materials more valuable and induces firms to install that equipment—that is precisely the point. But it does not do so as directly as regulation that forces firms to reduce emissions below a particular level or install a particular type of equipment. Any given firm would be less certain that regulated entities would install its products. Those firms will then have a smaller incentive to lobby for regulation, and might instead advocate for particularized command-and-control regulation as a substitute for Pigovian taxes.

In addition, Pigovian regulation might in many cases be considered inferior by regulated entities because it does less to inhibit new market entrants. There is a voluminous literature documenting the ways in which existing firms can use regulation to block competitors.160 Generally speaking, regulation can impose significant upfront costs on doing business, as when firms are required to purchase an expensive piece of equipment or obtain a costly or complicated permit. Large, existing enterprises may be able to surmount these regulatory hurdles, while new entrants might have much greater difficulty.161 Pigovian taxes do not have this feature, however. The cost of complying with a Pigovian tax scales linearly with the externality imposed; there is no initial upfront cost and thus no especially great hurdle to new entrants. Conditional upon being regulated, then, existing firms might in some contexts prefer conventional command-andcontrol regulation.

One concern about Pigovian taxes is that they may be difficult to administer. If the regulator cannot reliably determine the social cost of an activity, it cannot calculate the optimal Pigovian tax. This problem would of course affect Pigovian taxes whether imposed by statute or regulation. Economists have discussed these issues for decades and the consensus appears to be that administrative problems are generally manageable.155 In a recent paper, Victor Fleischer argues that Pigovian taxes are frequently inefficient because they are typically calculated based on an assumption of uniform marginal social cost when in fact the social cost of an activity often varies across individuals or firms.156 He provides the example of taxes on fatty foods, which may be justified on the theory that obesity creates social costs. However, some people eat fatty foods without becoming obese. They should not be taxed; and a tax might deter them from a valuable activity. A tax based on the average contribution of fatty food to social costs will also underdeter obese people. A better tax would vary according to the likelihood that any person becomes obese, but such a tax would be difficult to administer.

Regulatory nukes also have their corollary to strategic limitation and disarmament. 34  We have seen that very often a regulatory nuke that seems light anything more than a paper dragon faces two fates:  we either find a way to tame the beast or we kill it.  Limitations in this context can come from a number of sources.  This Part discusses three of them:  Congress, agencies themselves, and regulated entities.

3.6.1 Strategic Limitation By Congress