Thoughts for March 4, 2023

Good evening. This week’s topics are the aid debate, nuclear shipping, and kurtosis risk.

---

The Aid Debate

To this point, I haven’t dealt seriously with the question of whether Western aid to Ukraine, in fighting the Russian invasion, is well-advised. The main problem is that the anti-aid position have been dominated to pro-Kremlin and anti-Western voices which, frankly, I don’t consider worthy of serious response. But there are some well-thought anti-aid positions, and this essay, by Philippe Lemoine, is one of them. “Well-thought” doesn’t necessarily mean convincing, but there are some good points, and some issues worthy of discussion. Lemoine’s essay is long, but I would suggest reading it first so there will be context for my response.

Lemoine sets up and follows a table of contents.

Table of Contents

1. It’s extremely unlikely that, had the West not helped Ukraine, Russia would have attacked a NATO member next

2. Western military assistance to Ukraine makes proliferation more, not less, likely

3. Providing military assistance to Ukraine is not cheap once you take into account the indirect costs

4. The argument that committing to Ukraine’s defense was necessary to deter wars of aggression is flawed

5. The argument from credibility is a self-fulfilling prophecy and a recipe for the sunk cost fallacy

We’ll go through this section to section. On the first, Lemoine’s argument is the opposite of, say, that of Garry Kasparov, who argues that Russia under Putin is a fundamentally adversarial force to the West, like ISIS or Al Qaeda, and will do as much damage as they are permitted to do. Lemoine is probably right that a Russian direct attack against a NATO member is, under any circumstance, highly unlikely, but this misses some important points.

I discussed last year the doctrine of Russian doctrine of hybrid warfare and how this differs from the Western conception of war. Americans expect (though don’t always practice) a binary state of warfare: either two countries are at war or they are not, and wars end with well-defined peace treaties. The Russian doctrine of hybrid warfare envisions low-level, multi-decade, limited warfare that can be compartmentalized from other relations. In this regard, Russia considers itself at war with the West, including the United States, and has considered this for years. Examples of hostile actions under the banner of hybrid warfare include interference with the 2016 U.S. presidential election, the Solarwinds hack, and the Colonial Pipeline attack, among others.

There is also the threat of attack against further non-NATO states. Russia is backing separatists in the Transnistria region of Moldova, a move that bears striking resemblance to their backing of separatists in Luhansk and Donetsk in Ukraine. Lemoine is, I think, overly dismissive of the possibility that this will be followed by a full-scale attack.

On the second point, Lemoine argues that deteriorating relations between the United States and Russia over Ukraine aid jeopardizes arms control, as seen for example in Russia’s recent withdrawal from New START. I’m not too enthusiastic about arms control to begin with, and if looking the other way on imperialism is the price that must be paid to retain arms control, it really must be asked if that’s worth it. He also argues that deteriorating relations are inducing Russia to be bolder in assisting Iran’s ambitions to develop nuclear weapons and otherwise in selling weapons.

Here Lemoine contradicts himself. He argues at one point that Russia is not a serious military threat to the West because of their smaller defense spending, smaller GDP, and inability to conquer Ukraine. But he also argues that Russia is a potent threat through selling weapons. Here I would assess that he is wrong about the former and right about the latter.

On the third point, whether aid is “cheap” for the West is a calculation that would require more rigor than I am prepared to do now. But I would reiterate that American defense spending is near post-World War II lows and not adequate for the set of national security hazards, as well as too heavily oriented toward capability at the expense of capacity. Insofar as aid is expensive enough to pose a hazard, this is less an indictment of aid itself as more an indictment of under-prioritizing defense.

The fourth and fifth points are fairly similar, and I would regard them as the strongest that Lemoine has to offer. There are a number of would-be aggressors in the world today: for example Serbia against Kosovo, Azerbaijan against Armenia, Tajikistan against Kyrgyzstan, and China against Taiwan and India, though it is not all clear how the aggressors’ calculus would be affected by a strong vs. weak Western response in Ukraine.

Lemoine only discusses toward the end the moral revulsion that motivates the pro-aid sentiment. Aside from the fact that the invasion was entirely unprovoked, it has been characterized by widespread torture, rape, looting, and genocide. I will refrain from graphic portrayal of these war crimes, but I find it hard to understand how a person could be aware of them and not have an offended conscience.

Of course, there are many injustices in the world today, and it is not possible for an American-led NATO to respond adequate to all of them. And so there is a need to prioritize and exercise some hard-headedness in doing so. I place aid to Ukraine high on the list of priorities in large part because the war is being conducted by a country that has expressed open hostility to the United States and other Western powers, has attacked the United States in the past, and shows a clear desire to do so to the extent that it is permitted.

The difference with the isolationist view is that isolationists see the West, not Russia, as responsible for this. They hope that by burying their heads in the hand, as the ostrich proverbially does, the threat will go away. I think it’s pretty clear that this is wishful thinking.

Nuclear Shipping

There is a recent analysis on the use of nuclear power for shipping. We are familiar with the use of fission reactors for naval purposes—nuclear submarines, aircraft carriers, and ice breakers—and there is no technical reason why this couldn’t be done for cargo ships as well, though there are economic, logistical, and political challenges.

Under the Jones Act, any cargo ship between two American ports, including in Puerto Rico, must be built, owned, and operated by U.S. nationals. I regard the Jones Act as a costly protectionist measure that should be repealed, but that’s not what this analysis is about.

The number of Jones Act vessels, legally required for intrastate trade in the United States, is falling. From The Maritime Executive.

The decline in the number of Jones Act vessels is shown to be harmful to Puerto Rico in particular, creating serious headwinds to their hurricane recovery. And so the Maritime Executive analysis sees nuclear shipping, with (they consider to be) lower costs as a solution.

Four nuclear-powered cargo ships have been built worldwide. The United States built one, the NS Savannah, which served diplomatic functions and was intended to be the start of a fleet. The economics didn’t work out very well, though I would question how much generalization is possible from a single demonstration project. The article also highlights the problems that nuclear vessels have had docking.

Another issue would infrastructure for fueling and maintenance. Just a few ships probably wouldn’t work; a fleet would be necessary so ports would be equipped with facilities for waste disposal and refueling. I wonder if Jones Act ships, on which the analysis focuses, would be a sufficiently sized market.

The shipping industry is interested in decarbonization, though. The International Maritime Organization has set a goal of reducing the emissions intensity of shipping by 40% in 2030 and 70% in 2050, relative to 2008 values. This report, which as far as I can tell is typical of industry thinking, identifies alcohol (ethanol and methanol), biomethane, and ammonia as the most promising options. They don’t seem to put much emphasis on nuclear reactors. Most likely, if nuclear power is used in shipping in the late 21st century, it will be indirect insofar as synfuels can be generated on land from nuclear power.

Kurtosis Risk

In statistics, kurtosis is the standardized fourth moment of a probability distribution. Kurtosis risk is the risk that arises from a model whose kurtosis is greater than that of a normal distribution, as it typically assumed. In plain terms, kurtosis risk is also known as fat tail risk, referring to a situation where the long tails of the risk distribution are greater than that of a normal distribution. The mathematician Benoit Mandelbrot did some important pioneering work in the context of assert pricing. Kurtosis risk is a general concept which I think is important to understand to make sense of the world, even without an understanding of the mathematical details.

In the world on finance, Long-Term Capital Management has become the poster child for inadequately accounting for kurtosis risk. LTCM was a hedge fund founded by Myron Scholes and Robert Merton, both of who, would shortly receive a Nobel Prize in Economics for development of the Black-Scholes model of asset pricing. LTCM generated strong returns for a few years, but was undone by the Asian Financial Crisis in 1997 and the Russian Financial Crisis in 1998.

Several distributions exhibit high kurtosis, including the power law. For some exponents, power law distributions can have infinite mean, which is a highly counterintuitive property. Since any observation of a random variable from a power law distribution is finite, it means that the true mean is always greater than the observed mean, even over a very long observation period. Even in cases of finite mean, a power law distribution indicates that the true risk is usually greater than the observed risk over a period of time.

For example, Richardson’s Law holds that the severity of armed conflicts, later generalized to include terrorist attacks, follows a power law distribution. A consequence is that claims that international warfare has declined since 1945 might be more a statistical quirk than a genuine change in human society. Our World in Data also claims that public concern about terrorism is exaggerated, but the nature of kurtosis risk suggests that the actual danger is greater than the observed averages, from which they make their argument.

Peter Thiel likes to discuss the power law in the context of venture capital, noting that the returns of a venture investment tend to follow a power law distribution. 1% of all investments may yield greater returns than the rest combined.

Kurtosis risk has been argued to apply to damages from climate change, infectious diseases, and many other areas. For this reason, there is some justification in limited application of the precautionary principle toward risk management.