Crimea through a game theory lens

That is my latest NYT column and you will find it here.  Here is one excerpt:

Long before Malcolm Gladwell popularized the concept [of tipping points], Mr. Schelling created an elegant model of tipping points in his groundbreaking work “Micromotives and Macrobehavior.” The theory applies to war, as well as to marketing, neighborhood segregation and other domestic issues. In this case, the idea of negotiated settlements to political conflicts may be fraying, and the trouble in Crimea may disturb it further, moving the world toward a very dangerous tipping point.

First, some background: With notable exceptions in the former Yugoslavia and in disputed territories in parts of Russia and places like Georgia, the shift to new governments after the breakup of the Soviet Union was mostly peaceful. Borders were redrawn in an orderly way, and political deals were made by leaders assessing their rational self-interest.

In a recent blog post, Jay Ulfelder, a political scientist, noted that for the last 25 years the world has seen less violent conflict than might have been expected, given local conditions. Lately, though, peaceful settlements have been harder to find. This change may just reflect random noise in the data, but a more disturbing alternative is that conflict is now more likely.

Why? The point from game theory is this: The more peacefully that disputes are resolved, the more that peaceful resolution is expected. That expectation, in turn, makes peace easier to achieve and maintain. But the reverse is also true: As peaceful settlement becomes less common, trust declines, international norms shift and conflict becomes more likely. So there is an unfavorable tipping point.

In the formal terminology of game theory, there are “multiple equilibria” (peaceful expectations versus expectations of conflict), and each event in a conflict raises the risk that peaceful situations can unravel. We’ve seen this periodically in history, as in the time leading up to World War I. There is a significant possibility that we are seeing a tipping point away from peaceful conflict resolution now.

Do read the whole thing.

More generally, here is a new edited volume on the economics of peace and conflict, edited by Stergios Skaperdas and Michelle Garfinkel.

And here is the new forthcoming Robert Kaplan book Asia’s Cauldron: The South China Sea and the End of a Stable Pacific.  I have pre-ordered it.


broken link? your new york times article on Crimea is sending me to an economist article on Bitcoin

And a poor piece at that. As a 2 year resident of Yap the piece is silly.

On the other hand, remembr your organization theory. Have our foreign-military agencies ever said, "You know, threats are small, possibly receding. There are zero threats to American sovereignty, and almost zero to global trade. Some historic adversaries have imploded, like Syria and Egypt. Russia is a faint glimmer of its ugly self. Iraq is a zoo. N. Korea is a fifth-rate craphole, with about 1/20th the capabilities of S. Korea.
Iran? They exhausted themselves in a long war with Iraq and that was a draw."

Outside of internal conflicts (Syria), I can't think of any wars, unless you count Crimea. Maybe in Africa.

The most surprising thing on that list, to me, was that the Mexican drug war may have been the second most deadly conflict of 2013.

In general, there has been a salutary prejudice that the borders existing at the end of 1945 should be conserved; and when the borders do change, the change should not make big countries bigger.

The most significant exception to this was German reunification. The Russians are convinced that West Germany and the U.S. promised Gorbachev no expansion of NATO eastward in return for removing their 380,000 troops in East Germany. They may have a point:

In general, the world could get a lot worse than it has been since the end of the Cold War. Statesmen should not behave as if there could be no serious consequences to their recklessness.

In general, is letting small nations get smaller better than big nations get bigger? I'm not sure.

Correct link

The link points to an article about Bitcoin.

'the shift to new governments after the breakup of the Soviet Union was mostly peaceful'

The border settlements weren't always peacefult, though. Such as in the case, where the boundaries of a post-Soviet nation were altered through military means - - 'The Nagorno-Karabakh War, known as the Artsakh Liberation War in Armenia and Nagorno-Karabakh, was an armed conflict that took place in the late 1980s to May 1994, in the enclave of Nagorno-Karabakh in southwestern Azerbaijan, between the majority ethnic Armenians of Nagorno-Karabakh backed by the Republic of Armenia, and the Republic of Azerbaijan. As the war progressed, Armenia and Azerbaijan, both former Soviet Republics, entangled themselves in a protracted, undeclared war in the mountainous heights of Karabakh as Azerbaijan attempted to curb the secessionist movement in Nagorno-Karabakh. The enclave's parliament had voted in favor of uniting itself with Armenia and a referendum, boycotted by the Azerbaijani population of Nagorno-Karabakh, was held, whereby most of the voters voted in favor of independence. The demand to unify with Armenia, which began anew in 1988, began in a relatively peaceful manner; however, in the following months, as the Soviet Union's disintegration neared, it gradually grew into an increasingly violent conflict between ethnic Armenians and ethnic Azerbaijanis, resulting in claims of ethnic cleansing by both sides.'

And then there remains the interesting point of who Kalingrad/Karaliaučius/Królewiec (Königsberg is probably not in the running, even though it was the city's name before Kalingrad) really belongs to. Especially considering that current Russian rule is based on the same principle as Russian rule in the Kuril Islands - conquest by the Red Army.

Is it obvious that tipping points are (going to be) involved? It seems possible that even as peaceful resolutions become less effective, the decline of peaceful resolutions is linear or polynomial, not explosive (tipping).

Isn't it interesting that during this period of peacefully negotiated settlements the USA has been the one nation which has been constantly at war ?

Not really. The US was exceptional in many ways, including having the largest economy and military.

First off, excellent post today.

Second, I've been fascinated by the Putin - Obama dynamic for a while now. Obama has invested a lot of time deliberately shrinking the role of the US in world affairs. It has not been a coherent and well thought out policy, just the result of him being either disinterested or afraid. On the the other hand, Putin has been investing a great deal of time poking Obama and the US with a stick. The Left flushed the Syria debacle down the memory hole, but it is gives us a clue as to what comes next. Putin seems to relish sticking it to Obama so another NYTimes Op-Ed from him is likely.

Third, I think the proper way to game this out is to think about who is willing to absorb losses. This is not a winner take all battle. It is about who is willing to absorb unpredictable losses in order to win the day. Putin has set some limit above zero he is willing to absorb. The West faints when the DOW has two bad days in a row. Any act that has present economic costs exceeding benefit is now immoral.

It is not hard to see why Putin keeps winning.

What about Great Britain? They also signed that treaty on defending Ukraine territorial integrity.

Syria was Russia's client state...and Syria just self-immolated...this is a Russian victory? Syria could be ungovernable for decades...

Winning what exactly, vis a vis the USA? Let them.poke NATO or Japan and then I will consider that worth a fight.

Putin is a schoolyard bully taunting the world about his undisputed control of the distant rear of the playground, covered with broken glass and cigarette butts.

"Putin is a schoolyard bully taunting the world about his undisputed control of the distant rear of the playground, covered with broken glass and cigarette butts"

Outstanding analogy.

Obama haters are using this stuff as one more cudgel to beat him with. Even though staying out of Syria, and Ukraine, is EXACTLY what we should be doing. If Romney was doing the same thing Z would be all over his jock for careful statesmanship, and the liberals instead would be howling.

Syria is/way a Russian victory in that pledged western intervention was averted.

That was a US victory too. Maybe some don't see it but us not getting involved militarily in Syria is a win for the US.

And how much loss should we be eager to absorb in order to keep the Crimea Ukrainian? You seem to regard a sober assessment of our interests as unmanly--typical for someone who sees statecraft as a story of two kids on a playground. So tedious.

As far as deterrents go wrt Ukraine:
1. Nuclear is a non-starter, even if Ukraine had kept its nukes. A necessary factor is willingness to use it, and in its case its very close to zero with any arbitrary precision (even with conventional weapons not a shot has been fired even by its elite troops). Iraq had chemical weapons, it could probably deliver them to Israel and Kuwait -- didn't happen. Serbia when attacked could have mount strikes on Italian Aviano -- didn't happen. Weapons without will are useless.
Another factor is that even if used, it would most likely be by Russian PRO.

2. Market: not sure how it works. From what I have heard, 20+% stock market drop has been seen as a buying opportunity for the state to increase its stake in many important companies. Given Russia has cash reserves and can sit it through, not sure why it is bad for it.

We should restart the CPGS programs, and more generally build the means to threaten the Russian nuclear arsenal with conventional weapons, and a missile defense system that could deal with 50 or so surviving weapons. Ukraine was a clear illustration that our nuclear arsenal, which is ethically unusable except in the most extreme situation, provides very limited leverage. It's not much more useful than a handful of weapons in the hands of a rouge state, perhaps less so because of our reputation. We need to be able to threaten Russia without pretending we are going to initiate Armageddon, because everyone knows we won't do that.

Poincare ,studied basins of attraction within dynamical systems, which are the dynamical counterparts for looking at separatrices that can define tipping points. However, he did not do game theory, that being done by Borel prior to von Neumann, although I think neither did multiple equilibria when they did their game theoretic formulations. Proving existence itself was the big deal, with v-N's paper on this in 1928 being the first use of a fixed point theorem for proving such existence, to be extended later to proving existence of competitive general equilibrium. Study of basins of attraction and also the math of tipping points preceded Poincare.

So, this morning it looks like Russian troops have taken some territory beyond Crimea that has electricity and other supplies for Crimea. Probably can get away with that. The bigger question is if they go further, either to conquer eastern Ukraine or go all the way to Kyiv to reimpose Yanukovich, whom Putin continues to claim is Ukraine's legit president.

Nobody will invoke this just for Crimea, but there is the unpleasant matter of the 1994 agreement whereby Ukraine gave up its nuclear arsenal (to Russia) in exchange for a promise signed by Russia, US, and UK (and later France and China, clearly an agreement among the leading nuclear powers) to guarantee the "territorial integrity" of Ukraine. This is already violated, although clearly Crimea is a special situation. But a full assault on the rest of Ukraine surely brings this up, and clearly the US is the only nation that can do anything. Probably we will not, but Putin would be making a big mistake if he does so. US could pretty easily sink his fleet in Sevastopol if he overdoes this, although I do not think anybody can directly stop him from gobbling up as much of Ukraine as he wants to.

Oh, I cannot resist making a minor academic point. While Tom Schelling's presentation on tipping points in 1978 is very well done and widely known, he was not the first to analyze this, particularly with regard to residential neighborhood racial tipping. I beat him to the point in my PhD thesis in 1976 at the UW-Madison, although I did an odd formulation using moving boundary problem differential equations, probably the only such app of this in economics ever. Paper from diss published on this in Urban Studies in 1979, but largely ignored. I discussed the model in my 1991 book and again in my 2011 one. Neither I nor Tom had seen each other's work on this at that time

Of course, he had a different model involving nearest neighbor effects in his widely praised and seminal paper in 1971, which is the basis of modern agent based modeling, and which I reprinted in my 3-volume Complexity in Economoics book, praising it highly. However, I note that it does not analyze tipping points per se..

Goodness me, I'd written a comment featuring Poincaré but it doesn't seem to have shown up here. Great minds, eh?

Oh the embarrassment - I seem to have posted it on another thread. Or is that the NSA at work?

Don't be embarrassed, dearieme. I was actually replying to you but did so here. Should have done it there.

Oh, that 1971 paper by Schelling appeared in originally in the Journal of Mathematical Sociology.

This game theory ignores the reverse, based on human psychology. The more rare violence is, the more powerful even just the threat of it is. So the more peaceful we all become, the harder it is to deal with anyone who can credibly threaten real violence. Thus, the more likely it is that people (and nations) will threaten real violence. Think of it as cyclical. We may have reached the end of a peace-trending cycle as politicians around the world learn that they can overthrow governments, invade neighbors, have military coups and the world will at worst send them a weakly worded letter of concern. We won't even call it a "coup" when the military takes to the streets to remove a democratically elected government. Did we think no one would learn a lesson from that?

My thoughts exactly as I read this post. Also possible confusion between tipping and inflection?

The run-ups to the American Civil War and World War I are good counter examples. The publics unfamiliarity with the savagery of war increased their appetite for it.

And the same with the Crimean War, where British and French war tourists would sit atop the battlefield and picnic as the battle unfolded. Though, I do think apocalyptic nuclear war is an image still prominent enough in our culture to make the prospect of even a limited war between great powers unappealing to the public.

For that matter, cable TV had great ratings when the US was bombing the crap out of Baghdad. On the other hand, there was great disappointment in "Shock & Awe" when it was not a spectacular light show in cable TV.

But, the people don't have a say in any of this so it really does not matter.

That's not a counter example, it's a complementary example. It is a parallel process that acts in the same way as the one I describe. Too much peace increases the risk of war.

What's my investment play?

Go long Eastern Euro defense contractors.

One word: plastics.

Empirical analysis of determinants of conflict do not tell us amazingly much.

But the connection where an incidence of conflict is highly linked to increased probability of more conflict is fairly consistent.

Realistically, Russia has a fairly strong historical claim. We should be concerned with due process being followed in Crimea, not whether or not it ends up in Russian hands.

Does anyone deny the right of Porto Rico to ask to join the USA, for example?

If Russia has a fairly strong historical claim to any country where it once ruled and where some Russian population lives, they have claims to everything east of Munich.

Does anyone deny the right of Laredo Texas to join Mexico? Maybe after the Mexican military crosses the border and occupies the city?

Why predict the future just with speculative theory, while contemporary history and knowledge about the area provides so much more? Just remember Saakaschwili. Ukraine hasn't been part of any western alliance yet, Russia saw itself as Ukraine's hegemon. So did the west. Ukraine can break free of russia's hegemony, but it has to give up Crimea for this. Russia's game is to provoke Ukraine's nationalists so further military action becomes feasible, allowing Russia to occupy Oblasts like Donezk and Luhansk. If the Ukrainian nationalists are lured into this trap of russian aggression, they'll lose a lot. They'll lose more territory, more integrity of their country and the support of the west.

I've been involved with American military efforts to control WMD in the former Soviet states (including, frequently, Ukraine), and I keep seeing statements like Professor Cowen's "Yet if Ukraine were a nuclear power today, it would surely have a far greater ability to deter Russian military action." I think that's a hard counterfactual, but it's highly doubtful. If those states were still holding nukes there would have been a totally different relationship with the West -- you wouldn't be sending me and people like me over there to build scientific and technical infrastructure for one thing, but more importantly these wobbly nuclear states would've been seen as a threat and we might have encouraged rather than condemning Russian hegemony. Nuclear weapons haven't prevented a number of annexations and incursions; this case might have been different but I doubt it. What do you suppose the response of the West would've been if Ukraine -- which, again, would already probably be viewed suspiciously -- threatened to respond to Russian annexation of majority-Russian Crimea with a nuclear strike? Against whom would the majority of Western opprobrium be directed -- Russia, or Ukraine? I think Professor Cowen might have it right -- but he probably doesn't.

Good point; the West would probably have been eager to see Russia topple and de-nuclearize a state that, both in that hypothetical world and the real world, would have been wobbly, corrupt, and riven with pro- and anti-Russian internal conflict.

Good point. I don't see a lot of sentiment to the effect: "If only President Zuma still had the six nukes Mandela dismantled, the world would be a better place."

That's an awfully bad (and Sailer-esque: c'mon) analogy since South Africa _is_ the regional power; it isn't really relevant to Professor Cowen's idea that the liitle power Ukraine might need a defense against the bigger regional threat and hegemon. My thinking wouldn't really work there either; I don't know that it would be _as_ concerning to us that South Africa had nukes (we wouldn't be thrilled obviously) as it would be if all those regional powers in Eastern Europe that Andy Weber helped covince to disarm, hadn't. I suppose in theory you might worry about somebody nuking Lesotho but if it didn't happen under Mandela it's not real real likely to have happened since.

Honestly, Sailer, sometimes the racism descends into self-parody. A nuclear South Africa is a hell of a lot more like Russia than it is like a nuclear Ukraine.


Unfortunately only sometimes. Other times the racism is just racism.

Very good column, Tyler. The most interesting part, I think, 'market deterrence' wasn't in your excerpt. I'm not as much a '1914 pessimist' as you, but it was thoughtful and thought-provoking.

Some excellent comments too. This post sorta captures this blog at its best- unlike, say, open boarders trolling, which brings 'em out of the woodwork.

Keep on truckin'!

For a split second, I thought I read: "Crimea through a game of thrones lens."

It was, when Yulia was younger and more attractive

You might like question 2 of my PhD game theory midterm. It models the Crimean situation:

Comments for this post are closed