Monday, January 20, 2014

Does America's survival mean that it's a resilient country?



A friend asked me an interesting question yesterday: If a country survives a long time, does that mean we should expect its institutions to be resilient to future shocks? For example, American democracy has been around for over two centuries now, and has survived wars and depressions and ideological movements that have brought down the regimes of most other nations. Doesn't this mean that America has proven itself to be especially resilient?

The answer is: Maybe. It depends on what causes countries to topple.

Which is more reliable, an airplane fresh off the factory floor, or an airplane that has been in service for 30 years? Which is likely to be sturdier, a wall that has withstood 2 cannonballs, or a wall that has withstood 200? Which is likely to be more vulnerable to injury, a 20-year-old basketball player who has never been injured, or a 40-year-old player who has never been injured?

Let's think about this a little more concretely.

For example, suppose that countries get hit by "shocks" - wars, depressions, political crises. And suppose (for simplicity) that each country has a probability p_c of collapsing in response to any particular shock. p_c is a constant, but when a country begins, we have no idea what p_c is. We start with some prior distribution that represents our guess as to what p_c might be. As we see the country survive more and more shocks, we update our belief about the value of p_c. And, in keeping with my friend's hypothesis, the more shocks we observe, the smaller our guess of p_c will become. In other words, the longer the country survives, the more resilient we will expect it to be.

But that's not the only possible model for country survival. Here's an alternative: Suppose that a country has a set number of "hit points" - in other words, it will die after being hit with some number of shocks T_c. When the country begins, we don't know what T_c is, so again we start with some prior distribution. As we observe the country survive repeated shocks, our belief about T_c will change, and it will (weakly) grow - in other words, our guess about the total number of shocks the country can withstand will get larger and larger as we see the country survive more shocks. BUT, as more shocks occur, the country is also getting closer and closer to T_c - it's "hit points" are being depleted. Which effect is larger? It depends on the prior distribution. For example, if we start off believing that T_c is uniformly distributed between 1 and some number N, then our guess about T_c increases only at half the rate that the total number of shocks increases. Thus, the expected resilience (survival time) of the country goes down as the country withstands more shocks.

Why would a country be able to survive only a certain number of shocks? Well, a country might survive each shock by developing some sort of new institution - a regulation, a bureaucracy, a coalition of interest groups, etc. Those institutions might survive long after the shock has passed, and might make the country less efficient and less resilient to future shocks. This seems kind of like what Mancur Olson was thinking of. For example, America survived World War 2, but to fight the war we built up a huge military-industrial complex that might weaken our economy in the long run. We survived the Enron scandals, but we did so by implementing the Sarbanes-Oxley law, which may have severely hurt our capital markets in the long run.

Another even simpler model is "old age". Suppose that as countries age, the original problems that they were good at solving become less relevant, and new problems that they are not good at solving become more relevant. For example, the Qing Dynasty was great at subduing South China while simultaneously conquering Central Asian horse nomads. But when faced with the industrial revolution and the threat of seafaring empires, they collapsed. Again, "sticky institutions" mean that countries may have limited lifetimes. In math terms, take the "p_c" model, but assume that p_c is a decreasing function of time. If that function decreases faster than your belief about p_c(t) increases, then expected resilience goes down as a country survives longer.

There are a lot more, and more complicated models, but you get the idea.

(How about looking at data? What's the expected survival time of a country that has already survived 238 years? Unfortunately even this simple-seeming kind of exercise involves lots of assumptions, since different past countries might be more or less comparable to America - you don't want to lump Attila the Hun's empire in with the British monarchy. Also, there's the Industrial Revolution to contend with, which probably caused a structural change in the nature of the shocks that hit nations, and which came relatively recently in history.)

So to sum up: America's long lifetime might indicate that we're very resilient and durable, or it might indicate that we're due for a fall. Of course, I hope it's the former, but we really don't know. It all depends on what actually causes the rise, decline, and collapse of nations. And I don't think we know that yet.


Update: It has come to my attention that the argument that past longevity predicts future longevity was made by Nassim Taleb in his book Antifragile, and was referred to as the "Lindy Effect". I certainly hope Taleb did not attempt to use this principle to pick stocks, since this would constitute the Hot-Hand Fallacy.

34 comments:

  1. There is a little problem with the first hypothesis: if civilizations are more resilients over time, the consequence is that only a few civilizations will be destroyed and that will happens only to the newer civilizations. But we have a long list of civilizations that died and that lived long time before died: egyptian, romans, sumerians, maya, astecs, inca, byzantium, hitites, etc.

    So, aparently the historical data show that the first hypothesis is wrong. Adn the bad thing about historical data is that it say that ALL civilizations will die sooner or later.

    ReplyDelete
    Replies
    1. Anonymous1:10 PM

      I think you are mis-reading the first hypothesis. It doesn't say that surviving a long time makes a civilisation more resilient, but that surviving a long time makes high estimates of resilience more probable relative to low resilience estimates. It's about what you can reasonable conclude about the unknown quantity p_c, not that p_c is changing.

      Suppose you ordered a crate of 1000 widgets. If you tested a sample of ten and nine of them fail, it would be more reasonable to conclude that the order was poor quality than to conclude that you just happened to be very unlucky. The actual quality of the widgets isn't changing, just your knowledge about the quality.

      Delete
  2. Nations fall in the face of internal and external threats. One of the advantages America has had is that they are geographically isolated from their rivals. No one is going to think they can conquer the United States any time soon. Compare that with France.

    America has had relatively modest imperial ambitions. Those ambitions have generally ended badly but in ways that the US could survive. Compare the imperial experiences of Japan, Germany, and France. England may have been the one empire that did not end in disaster. An expansive imperial adventure by the United States could bring disaster in the form of an outright defeat followed by an internal political collapse.

    A nation fails if its political institutions stop working. Political institutions stop working when twenty per cent of the population start actively opposing them. The hardening of ideological lines in the United States since 1980, the soft corruption that is endemic in the political process (leading to policies that favor the rich and disenfranchise the poor) are the biggest threat to the ongoing stability of the United States.

    When the bottom seventy percent conclude that politicians have no integrity and the current game is rigged against them, they will flip over the card table and demand a new game.

    ReplyDelete
    Replies
    1. Nathanael6:19 AM

      Yep.

      Delete
    2. Nathanael6:21 AM

      FWIW, this happened before. It was called the Civil War. And again -- it was called the New Deal. Arguably it happened in between as well, during the late 19th century union wars, with the Progressive movement, when there were wars of competing governments in several states in the Midwest.

      The New Deal settlement has been *remarkably* stable. The US was *extremely unstable* prior to that.

      Delete
  3. Freeman Dyson had a great little heuristic he used to estimate the longevity of the human species: there's a 90% chance we're living in the middle 90% of the lifetime. Therefore 238 years implies that the US will survive to between 264 and 2380, or between 26 and 2000 more years ... Which puts us in line with other civilizations.

    ReplyDelete
    Replies
    1. Try that strategy to predict runs of good stock returns, I dare you! ;-)

      Delete
    2. Ha! I think the strategy effectively tells not to try to make predictions on stock returns.

      Delete
    3. Really?

      Suppose we just saw stock returns be unusually good for 5 days. By this principle, it seems like we should expect that the good returns will continue for another 5 days. Isn't that correct?

      Delete
    4. This logic has also been applied to trying to puzzle out how long humanity will survive as a species. It makes as little sense there as here.

      The US has existed for 238 years. Assuming our entire society collapses tomorrow, the first 5% of its lifespan was its first 11.9 years, and the last 5% of its lifespan was its last 11.9 years.

      So our calculation that the US could live for another 2000 years is absurd, as is the calculation of John Adams that America would surely collapse within his son's lifetime.

      The issue is that the estimate produced by the calculation changes, wildly, based on when the calculation is being taken. Without any metric for determining whether or not one is, in fact, in that 90% range means that the whole exercise is basically sillyness.

      Also, why 90%? There's a 95% chance that we're in the middle 95%! But there's only a 50% chance that we're in the middle 50%. And so forth.

      Also, there have been a lot smarter rebuttals of this argument than mine:

      http://hanson.gmu.edu/nodoom.html

      Also, think about it this way - if you asked this question to every single dinosaur, then 90% of them would be correct to assume that they were in the middle 90%. But I bet if you were able to construct a roughly accurate time distribution of the dinosaur population, then substantially more than 10% of the dinosaurs would've been wrong, since the population likely increased continually over 135 million years until the K-Pg event.

      To summarize - that argument makes certain assumptions about the parameters of population distribution over time as well as one's own place in that distribution that cannot be supported axiomatically.

      Ah, snow days.

      Delete
    5. Anonymous12:14 PM

      "Try that strategy to predict runs of good stock returns, I dare you! ;-)"

      Haha I think we should start applying this to all theories, physics, biology, theology as well. We'll call it the Noah test. If you think evolution is a real try making money off the stock market with it I dare you. Think Shiites have the Truth, they should try making money of the stock market with it.

      Delete
    6. Look, stock markets are pretty short-run things compared to evolution. In the long run the Sea People will burn all our cities well before the stock market can generate meaningful alpha for belief in evolution. That doesn't mean that markets can't provide some meaningful feedback for theories that fall within the same time-scale as markets.

      Delete
  4. There's another model, which Noah hinted at when mentioning how the industrial revolution changed the nature of threats.

    That model can be analogized with evolutionary biology and the fitness of species. Over the eons, we've seen many different species (or rather, entire Orders) come and go, even just within the Animalia. Many of these were well suited for their environments at the time and persisted for tens millions of years (think dinosaurs).

    But eventually, the nature of their world changed (due to environmental changes and evolving competition) and they could not adapt, so they died out. New types of organisms came in to take their place which were more well adapted to the new world.

    Nations can be viewed this way. The ones that emerge and persist have shown themselves to be robust in the face of whatever challenges they have faced. That does not, however, imply they will necessarily be robust in the face of future challenges. As technology and societies evolve, we can expect the threat environment to shift. So a USA that is robust to the changes wrought by major wars and industrialization may or may not be ready for terrorists wielding nanobots and Skynet.

    Still, I think it's a bit presumptuous (even though it's intended as the opposite) to just throw hands up and declare "we don't know." One thing I notice is that not only has the United States persisted for a couple centuries, but so has just about every developed country that has adopted its model of small-L liberal democracy. That doesn't prove the ability of that model to survive any and all future shocks (think about all the different families of dinosaurs that ALL died out eventually), but it does broaden our data set to include countries that have survived more and more varied challenges over a longer set of collective years. Looks pretty robust to me.

    ReplyDelete
    Replies
    1. Nathanael6:17 AM

      Unfortunately, no other country has actually adopted the US's model of liberal democracy.

      The US has a particularly dysfunctional and messed-up "democracy" compared to most other countries in the world. The US Senate is example one. Also, the "Presidential system" seems to be bad overall, and countries with "Presidential systems" fall to military coups far more often than countries with "Parliamentary systems".

      So there does seem to be some form of adaptability associated with democracy -- but more with *parliamentary* systems than with systems like the US. Also, *proportional representation* seems to be more adaptable than first-past-the-post...

      The US has been unable to reform its system to become a modern democracy -- and people aren't even *talking* about such reform. This is a bad sign.

      Delete
  5. One could hypothesize that nation states are a relatively new phenomenon but that once they come into existence they create a situation that is difficult to change. For example: the German mini-kingdoms were probably more fluid than the current German state. And with modern communication (since the mechanical telegraph) states have better means to keep themselves together. And borders are better delineated. So one could hypothesize that modern states have much stronger internal cohesion than states in the 17th century. So any comparison might be spurious. - Compare this with the lifetimes of companies. There is much research on that. What are the patterns in there? (You economist probably know that better than I.)

    ReplyDelete
  6. so has just about every developed country that has adopted its model of small-L liberal democracy.

    Don't forget that the United States went through a civil war so it has not all been smooth sailing. Other countries became liberal democracies between about 1848 and 1920 but quite a number of those back slid into variations of fascism including Greece, Germany, Italy, Spain and most or all of Latin America. France has gone through multiple convulsions. Liberal democracy has not been as robust as you seem to think it has. Being predominantly Protestant seems to promote stability. :-)

    ReplyDelete
    Replies
    1. The US went through a Civil War ... and survived. Other countries had back slides only very early on after becoming democracies, but almost no modern democracies have ever gone back once established for any appreciable length of time, Catholic, Protestant or otherwise.

      So no, I think you are mistaken. I don't know the deep history of every nation (!), but I see that every country you list has decades of stability under it's belt (aside Latin America, which is outside of the "developed" criteria I originally stated) and there hasn't been a war among them in three generations.

      Even now, with an economic crisis comparable to the Depression in much of Europe, we are seeing ... not Hitler, but rather popular disgruntlement under stable governments. There's your sign.

      Delete
    2. Nathanael6:13 AM

      External interference prevented most Latin American democracies from stabilizing, FWIW. I've studied the history of that.

      A democracy needs to be left alone for a while in order to stabilize.

      However, retaining "democracy" is no guarantee of stability. Nowadays, almost all revolutions will claim to be creating a better form of democracy than the form which was present before. For an early example, look at the replacement of the French Fourth Republic with the French Fifth Republic.

      Delete
  7. 200 years is not really considered long, since the average life span of large empire is about 300 years. for example: sassanian empire 425 years, han dynasty 400 years, ming dynasty 250 years, ottoman empire 450 years, qing dynasty 250 years, Umayyad Caliphate 60 years, mongol empire 160 years, British empire 300 years

    Here's a prediction: since gold standard has been used for thousands of years, therefore one day nations will return to some form of gold standard.

    ReplyDelete
  8. This is a time when looking into the past probably does not help.
    500 years ago very few people could read and even fewer had a written history.
    Those people did not have the chance to learn from other civilisations / cultures around the world.

    Whether or not this knowledge will really help or not will become clear ove the next 1000 years or so

    ReplyDelete
  9. Then there is the little problem of defining "survive" and what exactly must survive to consider the abstract concept of a state, or society, or peoples to have survived. The Anasazi survive to this day in the form of the Pueblo Indians, but the great cities they built like Chaco survive only as museums. The "country" of Texas survived its independence from Mexico only for a short period of time before becoming part of the US, so did Texas survive? Genghis Khan's Golden Horde once ruled most of Asia and parts of the Middle East and Eastern Europe. That is no longer true but his ancestors still rule Mongolia. Perhaps "survive" is meant to imply a certain form of government, within a set of defined borders? So Rome didn't survive (lost it's form of government and empire-defined borders). The Soviet Union didn't survive for the same reason.

    Or are we really talking about extinction -- completely disappearing in every way except history books?

    ReplyDelete
  10. In shorter terms - what history seems to show is that, once established, democracies in developed nations don't fall.

    ReplyDelete
  11. Huh?
    For example, American democracy has been around for over two centuries now, and has survived wars

    The most deadly war for Americans, by a long shot, was the one brought about by the.. er, imperfections of American Democracy is the Civil War.

    and depressions and ideological movements that have brought down the regimes of most other nations.

    Which regimes were brought down by the Depression?
    Which ideological movements threatened Am-Dem? And are sure they weren't a product of rather than a threat to the system?

    American Democracy is less democratic and more dysfunctional than most advanced democracies. So much so that it is a fair question to ascribe any putative resiliency to its authoritarian qualities or to its inefficiencies as to any positive virtues.

    ReplyDelete
  12. "I certainly hope Taleb did not attempt to use this principle to pick stocks, since this would constitute the Hot-Hand Fallacy."

    You apparently failed to notice that Lindy effect only apply to non-perishable things.

    ReplyDelete
  13. Anonymous4:51 AM

    Please tell me that macroeconomic objects you and your colleagues normally consider are not as loosely defined as state/country/empire/civilization/[certain set of laws?]/[certain set of rulers?] that your post considers here. Likewise that survive/fail/[experience a superficial change in form of government or ruling class?] are not as loosely defined. Or that shocks/events/solving problems/failing to solve problems/resiliency are not so loosely defined. Tell me you don't apply math to metaphors. Please re-assure me, a newbie requests.

    ReplyDelete
  14. There is another problem with this: The definitions of a country and survival. As an example of this problem, take France and Britain. You can convincingly argue that France has existed since Clovis. But since the end of the 18th century France has had five republics, two empires, a restoration of monarchy, and two special regimes. And let's not even get into all the problems associated with the concept of "France" in the middle ages, with Charlemagne's empire, the division of the realm, that English/Norman-French spat in the High Middle Ages, and so on. France as a (rough) concept has arguably existed for a very long time, evolving since the land was called Frankia. But the country has changed to an incredible extent.

    Same with Britain. It has survived most shocks - but its institutions have changed so that the functioning of its constitutional system is radically different than it was in, say, 1800.

    What I'm saying is that this line of inquiry is dubious at best if the researchers don't agree what constitutes survival, which discontinuities qualify as collapse, and which qualify as adaptations.

    As an aside, under international law, a country is defined by three basic traits: It has a territory, a population, and state institutions are able to exercise effective (and independent) control over it. That might serve as a good starting point.

    ReplyDelete
    Replies
    1. Indeed, these are very good questions. But if a country loses control, then regains it, did it survive?

      Did England "survive" its civil war?

      Frankly, this conversation makes me think that survival is just a really bad framework for discussion macropolitical phenomena.

      Delete
    2. Nathanael6:10 AM

      The goal, in my opinion, is to have a system which can adapt without requiring a civil war.

      The US system has been bad at this. It failed completely in 1860. On the other hand, it succeeded in 1932. On the third hand, it seems to have been failing to adapt to change since 1980.

      Britain has been extremely successful since the end of the English Civil War. The institutions are practically unrecognizable, having changed massively, but they've managed to do so while avoiding a domestic civil war since then. They came close in 1832.

      France failed repeatedly.

      Delete
  15. Please define your terms.

    This is not snarky! Seriously! Here's what I'm getting at:

    Texas and Alaska secede permanently from the Union, forming the PetroStates of RealAmerica. Did the United States survive?

    The United States is present in 500 years, but is the most economically, culturally, and politically backwards nation on Earth, and is alternatively mocked, scorned, and pitied by the vastly more advanced and civilized societies in Africa and Central Asia. Did the United States survive?

    The democratic institutions of the United States collapse, and is replaced by a dynastic imperial authoritarian government, that nonetheless presides over cultural and economic flourishing over the same defined territorial borders for another thousand years. Did the United States survive?

    These questions matter! You put the picture of the Colosseum to illustrate the post, but there's no obvious consensus on when "Rome" as a political entity ceased to exist. We all gravitate towards 10476 HE/476 CE, but here are some perfectly good alternative responses:

    Rome ceased to exist in 9974 HE/27 BCE, when Augustus became the first emperor, putting an end to centuries of republican rule.

    Rome ceased to exist during the Crisis of the Third Century, when it fractured into three wholly independent empires; the political entity that followed Aurelian's reconquest, Diocletian's reforms, and Constantine's catalyzing of Christianization was a wholly different entity than the empire that ruled the Mediterranean for the past several centuries.

    Rome ceased to exist in 10476 HE/476 CE, when Odoacer deposed Romulus Augustulus - but note! He sent the purple ropes to Xeno, the Eastern Roman Emperor, and acknowledged him as sole ruler of all Rome, and that he was simply ruling in his name. And note that less than a century later, the Eastern Roman Emperor Justinian I reconquered North Africa, Italy, and part of Spain, and might have knit the whole kit-and-kaboodle back together, if it wasn't for that gosh-darn worst-pandemic-in-human-history.

    Rome ceased to exist in the 107th-108th centuries HE/7th-8th centuries CE when everything but a rump state mostly centered in modern Turkey, Greece and southern Italy.

    Rome ceased to exist in 11204 HE/1204 CE when Constantinople was sacked by the Latin Christian Crusaders and supplanted by the Latin Empire.

    Rome ceased to exist in 11453 HE/1453 CE, when the rump city-state formed after the destruction of the Latin Empire was permanently supplanted by the Ottoman Turks.

    Which of these is the right answer? Note that the empire that reigned in the Eastern Mediterranean for 150 years following 10476/476 contained the vast bulk of the Roman Empire's wealth, commerce, and culture - the Western Roman Empire whose fall we so elevate as a demarcation of history were the mostly poor and backwards provinces of the empire; the Eastern Roman Empire suffered far more economically from the loss of North Africa then it did the loss of Spain, France, and Italy.

    Just saying.

    ReplyDelete
    Replies
    1. Of course, you're completely right.

      Which is one more reason why attempts to mathematize history are not likely to succeed.

      Delete
    2. Yep.

      Or, at the very least, that attempts to more formally model historical movement need to be much more modest and well-defined.

      For example, you might be able to make a probability model that predicted "percentage of a given population that will die of violent causes in a certain time period and/or likelihood of substantial change in that percentage" based on prior factors, but you can't predict "civil war" or "country survival" without defining that suuuuuper rigorously and then you'll always have liminal cases whose inclusion or exclusion alter your conclusions that confound you mercilessly.

      Not to mention endogeneity.

      Delete
  16. Verily , It is a graceful and fundamental website about financial market. It is a very beneficial for us. So, I like it very surely. Many many thanks for make this website. If you want more informastion about stock quotes to visit stock quotes Stock quotes can get gathered inside numerous kinds just as historical stock quotes, delayed stock quotes and real-time carry quotes. Historical stock quotes are share values not to mention improve behaviours previous to several amount of time - employed to know not to mention pinpoint monthly carry trends.

    ReplyDelete
  17. Anonymous9:17 AM

    In any attempt to model the state collapse it seems to me that the mathematical models of evolution a' la Yule process are a way to go. Most of the civilizations collapsed while interacting with other civilizations so a process of copying (successful strategies) and mutation (adapting in other environment) coupled with varying fitness function could be the way to go. There is no clear evidence but if you can show a fundamental drift in fitness and a tendency of big-old civilizations to keep to their memes and avoid mutation perhaps you could actually model this process. At least such framework would allow to compare civilizations before and after industrial revolution although I do not have idea how you could quantify these fitness variations.

    ReplyDelete
    Replies
    1. Nathanael6:02 AM

      Anonymous is correct. Evolutionary models are the correct models.

      The English government was highly durable because it proved highly adaptable. Institutions were thrown by the wayside or "reinvented" on a regular basis to deal with challenges. England would be quite unrecognizable in some ways even to someone from 1910. The US... would be more recognizable, unfortunately.

      The ancient Egyptian government, by contrast, was highly durable only because it had found a niche where its environment did not change for thousands of years, so it didn't *need* to adapt. When the technological and social environment changed enough, it fell apart quite quickly.

      Delete