Intra-Elite Competition: A Key Concept for Understanding the Dynamics of Complex Societies

Peter Turchin


Join 36.9K other subscribers

Intra-elite competition is one of the most important factors explaining massive waves of social and political instability, which periodically afflict complex, state-level societies. This idea was proposed by Jack Goldstone nearly 30 years ago. Goldstone tested it empirically by analyzing the structural precursors of the English Civil War, the French Revolution, and seventeenth century’s crises in Turkey and China. Other researchers (including Sergey Nefedov, Andrey Korotayev, and myself) extended Goldstone’s theory and tested it in such different societies as Ancient Rome, Egypt, and Mesopotamia; medieval England, France, and China; the European revolutions of 1848 and the Russian Revolutions of 1905 and 1917; and the Arab Spring uprisings. Closer to home, recent research indicates that the stability of modern democratic societies is also undermined by excessive competition among the elites (see Ages of Discord for a structural-demographic analysis of American history). Why is intra-elite competition such an important driver of instability?

Elites are a small proportion of the population (on the order of 1 percent) who concentrate social power in their hands (see my previous post and especially its discussion in the comments that reveal the complex dimensions of this concept). In the United States, for example, they include (but are not limited to) elected politicians, top civil service bureaucrats, and the owners and managers of Fortune 500 companies (see Who Rules America?). As individual elites retire, they are replaced from the pool of elite aspirants. There are always more elite aspirants than positions for them to occupy.  Intra-elite competition is the process that sorts aspirants into successful elites and aspirants whose ambition to enter the elite ranks is frustrated. Competition among the elites occurs on multiple levels. Thus, lower-ranked elites (for example, state representatives) may also be aspirants for the next level (e.g., U.S. Congress), and so on, all the way up to POTUS.

Moderate intra-elite competition need not be harmful to an orderly and efficient functioning of the society; in fact, it’s usually beneficial because it results in better-qualified candidates being selected. Additionally, competition can help weed out incompetent or corrupt office-holders. However, it is important to keep in mind that the social effects of elite competition depend critically on the norms and institutions that regulate it and channel it into such societally productive forms.

Excessive elite competition, on the other hand, results in increasing social and political instability. The supply of power positions in a society is relatively, or even absolutely, inelastic. For example, there are only 435 U.S. Representatives, 100 Senators, and one President. A great expansion in the numbers of elite aspirants means that increasingly large numbers of them are frustrated, and some of those, the more ambitious and ruthless ones, turn into counter-elites. In other words, masses of frustrated elite aspirants become breeding grounds for radical groups and revolutionary movements.

Another consequence of excessive competition among elite aspirants is its effect on the social norms regulating politically acceptable conduct. Norms are effective only as long as the majority follows them, and violators are punished. Maintaining such norms is the job for the elites themselves.

Intense intra-elite competition, however, leads to the rise of rival power networks, which increasingly subvert the rules of political engagement to get ahead of the opposition. Instead of competing on their own merits, or the merits of their political platforms, candidates increasingly rely on “dirty tricks” such as character assassination (and, in historical cases, literal assassination). As a result, excessive competition results in the unraveling of prosocial, cooperative norms (this is a general phenomenon that is not limited to political life).

Death of Gaius Gracchus (François Topino-Lebrun) Source

Intra-elite competition, thus, has a nonlinear effect on social function: moderate levels are good, excessive levels are bad. What are the social forces leading to excessive competition?

Because the supply of power positions is relatively inelastic, most of the action is on the demand side. Simply put, it is the excessive expansion of elite aspirant numbers (or “elite overproduction”) that drives up intra-elite competition. Let’s again use the contemporary America as an example to illustrate this idea (although, I emphasize, similar social processes have operated in all complex large-scale human societies since they arose some 5,000 years ago).

There are two main “pumps” producing aspirants for elite positions in America: education and wealth. On the education side, of particular importance are the law degree (for a political career) and the MBA (to climb the corporate ladder). Over the past four decades, according to the American Bar Association, the number of lawyers tripled from 400,000 to 1.2 million. The number of MBAs conferred by business schools over the same period grew six-fold (details in Ages of Discord).

On the wealth side we see a similar expansion of numbers, driven by growing inequality of income and wealth over the last 40 years. The proverbial “1 percent” becomes “2 percent”, then “3 percent”… For example, today there are five times as many households with wealth exceeding $10 million (in 1995 dollars), compared to 1980. Some of these wealth-holders give money to candidates, but others choose to run for political office themselves.

Elite overproduction in the US has already driven up the intensity of intra-elite competition. A reasonable proxy for escalating political competition here is the total cost of election for congressional races, which has grown (in inflation-adjusted dollars) from $2.4 billion in 1998 to $4.3 billion in 2016 (Center for Responsive Politics). Another clear sign is the unraveling of social norms regulating political discourse and process that has become glaringly obvious during the 2016 presidential election.

Analysis of past societies indicates that, if intra-elite competition is allowed to escalate, it will increasingly take more violent forms. A typical outcome of this process is a massive outbreak of political violence, often ending in a state collapse, a revolution, or a civil war (or all of the above).

Notify of
Most Voted
Newest Oldest
Inline Feedbacks
View all comments
Gene Anderson

Works for China too. One can see two main sources: The Imperial family, which with vast-scale polygyny grew inordinately in a short time; and the examination system, producing more and more successful candidates over time (this was a problem mainly after Song greatly expanded the exams). The poor Imperial family deserves some pity–toward the end of a dynasty you had all these 13th cousins 10 times removed starving to death on the Russian frontier. (I exaggerate only slightly. By the end of the empire in 1911, there were tens of thousands of Imperial relatives.) Naturally the competition got pretty fierce late in the dynasties. When the empire thrived, the system could blot all these people up, and find places for them. When the empire was going down hill, or conflicted, it meant trouble.

Rick Derris

I liked the intra-elite discussions in “Ages of Discord” and it made me an even more strident believer in term limits. At least moving people out of the Congress after eight years will “free up” some space for other elite aspirants. I don’t care if your politics are on the side of Strom Thurmond or Ted Kennedy – both were in the Congress for far too long.

Of course, term limits did nothing to keep a 2nd Cuomo out of the NY Governor’s mansion, but at least it means we only have to watch one Cuomo on CNN.


I believe Peter Turchin is deeply mistaken about elite competition in modern societies. I repeat my comment on intra-elite competition from a previous post:

In an agrarian society, elite wealth was based on land, more specifically, on extracting a fraction of the output of the commoners working the land. When there was a demographic crisis (land-labour ratio fell and immiseration set in), elite incomes fell, and elites sought to maintain their lifestyles by increasing the rate of extraction. But squeezing peasants even more when there’s already a demographic crisis only exacerbates popular immiseration. At some point the only way for elites to increase, or even just preserve, their incomes was at the expense of other elites. Thus you have elite fragmentation and internecine competition. And thus sociopolitical instability. Makes a lot of sense. It fits a lot of historical cases.

However, this theory makes no sense in modern industrial societies.

(1) Wealth is no longer fixed in the long run. Modern economies reliably grow at 1-2% rates. Much of that growth is concentrated at the top, even when measured income inequality is relatively low. So the competitive pressure within elites is much less than in any agrarian society governed by Malthusian-Ricardian-Brennerian-Goldstone-Turchin cycles.

(2) Besides, in a modern society, you need *more*, not less, intra-elite cooperation (a) in order to increase economic inequality; (b) in order for the elites to capture a greater share of the economic growth; (c) in order for capitalists reduce the bargaining power of labour; and (d) in order for elites to capture the state.
In fact, politics in a modern society is a pretty small part of the field in which elites can play compared with anti-competitive practises — i.e., collusion, mergers, monopolies, trusts, and other ways of reducing competition and concentrating power in the supply of goods and the demand for labour. These are all acts of elite cooperation. Capitalists are, right now, in unprecedented unity. They agree on unions, immigration, wages, trade, regulations, etc. That unity is necessary to generate the inequality in the first place.

Therefore, state capture and rent-seeking are now *cooperative*: conspiracies to rig the rules and increase markups against the public interest require collusion. Owners of one mobile telephony operator don’t have to clash with the owners of another mobile telephony operator: they can band together to lobby the government. Compared with the rise of monopoly concentration, elites wrangling over Trump or Brexit is a sideshow.

Almost everybody who is concerned about rising inequality implicitly recognises this: from Krugman to Stiglitz to Milanovic to even Turchin’s friends at Evonomics, they have argued that inequality stems in great measure from anti-competitive practises.
It’s contradictory to bemoan the spread of the ‘neoliberal’ ethos, and simultaneously talk about elite fragmentation. The evidence Turchin marshalls for elite fragmentation is basically the bimodal distribution of lawyers’ incomes, and the degree of legislative polarisation. He ignores the much wider evidence of capitalist unity and concentration in support of ‘neoliberal’ policies.

Fernando E.Mora

I think you must read Fred Hirsch’s “Social Limits to Growth” to understand the difference between the always possible growth in MATERIALl wealth and the (no-)growth of POSITIONAL wealth in which Peter’s point can also be solidiliy (and perhaps more accurately) based.


I would certainly agree that if economic growth were zero or negative, PT’s elite competition theory might make more sense. Which is why I think SD theory is still quite applicable to many contemporary developing countries, such as those in the Arab world. Also, the collapse into civil wars in many African countries in the 1980s and 1990s was preceded by a large expansion of educated people at the same time economic growth more or less came to a halt.

Stephen Morris

Speaking as a former investment banker involved in the privatisation of public assets – who has seen at first hand generations of politicians captured by business interests – I suggest that anyone with direct experience of this matter would realise that any collective action problem faced by the capitalist class in negligible in comparison which the collective action problem faced by citizens under the non-democratic system of purely “elective” goverrnment (i.e. “government-by-politicians’).


Re #1 — No, I do not measure everything in money, so please do not write a whole post as though that’s what I argued. I said that elites now *collude* to capture the political process, which they do. They don’t need to compete for political positions because they cooperate in capturing it. Goldman Sachs has access to the Treasury department whether the party in power is Republican or Democratic. (Besides, you also use some money proxies for intra-elite competition/cooperation: the distribution of lawyers’ salaries, or the Great Merger Movement.)

Re #2 — I do not assume it. The evidence is overwhelming that concentration is increasing, markups are rising, monopoly power is expanding. All of that is evidence of intra-capitalist cooperation and unity.


Peter Turchin frequently cites the work of Martin Gilens, who has repeatedly shown that public policy largely reflects the preferences of the very richest of US society. That’s not elite competition. That’s elite cooperation in capturing of the political process. The problem with Turchin’s framework is that he sees even modern societies through the Roman framework of Optimates v. Populares.


pseudoerasmus, I pretty much agree with what you say. However, while elites have colluded to capture the political process we might not expect them to all agree on what to do with the political process once it has been captured.

There is no intra-capitalist unity. Some elites shouldn’t even be called capitalists because the monopoly power they seek completely eliminates the free market. Other elites who want to control the political process do want a free market. They are in conflict.

The common thread here is the presence of powerful elites who cooperate. Historically the monopoly power elites have cooperated without much resistence but the free market elites have begun to cooperate against them and have had success in the election of Donald Trump.

If it is people power we want then the general trend will look like cooperation as whoever wins the conflict will be cooperating economic elites.

Steve Roth

I question whether there is a qualitative difference today. It’s still about the claims embodied by “wealth,” and the power those claims impart to wealthholders. The mechanisms are different, but the wealth/power relationships are pretty much the same.

The crux, in my view, is concentration of wealth (hence power). Which has the virtue of being nicely quantifiable, in concept if not necessarily in practice.

My favorite graph of this:

As concentration increases and the “elite” gets smaller, the rope-ladder hanging down from the elite gets shorter and rattier. eg: The 90% were always excluded. Now the 2%-10% are. That change could result in a different type or intensity of social conflict.

On the other hand that intra-“elite” competition might just be a by-product and analytical distraction. The elite vs “the rest” is the issue, and all we need to look at is the size of the elite. That could be nicely encapsulated in a “wealth concentration” metric.

Problem is getting a consistent measure of that wealth concentration. Hell, the U.S. national accounts didn’t even tally wealth until 2006, and still don’t even touch on wealth distribution.

Assembling such a (validly consistent) measure across historical societies would be tough. Atkinson, Wolff, Piketty&Co, etc. have managed over recent decades to assemble data on richer countries going back a century or so. Perhaps one could do similar for the Roman Empire, at least roughly? But across many societies and millennia? Tough.


In agrarian societies, the wealth that conferred status — land and state offices — were fixed in the long run. In modern societies, the supply of status positions is not fixed and is in fact highly elastic.

Steve Roth

Yes the quantity of wealth was fixed. But I’m talking about the concentration of wealth and power. Compare a society in which the 1% has all the wealth and (real) power, compared to one where it’s more broadly distributed among the 10%.

IOW, whaddaya mean by “elite,” buster?

>the supply of status positions is not fixed and is in fact highly elastic

Totally agree. Increasing wealth does not mean that the quantity of status positions is increasing. The absolute or percentage count of “the elite” could shrink (wealth could concentrate) even as wealth increases.

Increasing wealth might be presumed to give more entree to aspirants than a fixed-wealth scenario, but I just have no idea whether that is actually the case.

Dick Burkhart

You claim that “wealth is no longer fixed in the long run”, yet that claim is the most fundamental fallacy of contemporary economics. “Limits-to-growth” is not a choice but a fact of science. Already the global economy is stagnating, mostly for this reason, and it is headed toward contraction sometime during the coming generation, despite all the hype about new technologies.

The concept of “ecological overshoot and collapse” applies to human ecology too. We’re certainly in overshoot, so some form of collapse is coming (even if a technological miracle occurred, like cheap energy from nuclear fusion, it would only postpone the day of reckoning).

As to “intra-elite competition”, it is well underway in much of the upper middle class and the 1%, according to the statistics documented by Peter Turchin above. But it is just revving up among the super-elites – the billionaire class, with Trump being the first really visible eruption. In fact, Donald Trump’s election is the perfect example of how this competition plays out once it hits the main stage. So don’t confuse tactical cooperation among increasingly greedy factions of the elites with the kind of yawning political fractures that are now opening up as unscrupulous opportunists like Trump discover that they can exploit a disgruntled part of the populace to “trump” the more conventional elites. And as “limits-to-growth” blocks the customary relief valve of expansion, then elite exploitation and popular revolt will increase until something there is some kind of show stopper.


Repeat after me: most of GDP isn’t material, and the material content of the extra unit of GDP is falling

Besides, even if ecological collapse *were* coming, current problems have nothing to do with anything ecological or resource constraints.

Dick Burkhart

Like most economists, you’ve got it totally backward: The non-material part is completely dependent on cheap resources, especially cheap, and compatible ecosystem conditions. Those resources only seem to disappear from the economy, because they are so cheap. But, as in the rest of nature, all that complexity comes from the surplus of energy and other resources.

After all, we could not live without good air. Yet it costs nothing most of the time, so doesn’t even enter into conventional economics.


Well, Dick Burkhart, as I said earlier, even if ecological exhaustion and collapse were coming, (a) that is not related to current economic problems; and (b) it’s also not part of Peter Turchin’s diagnosis.

Dick Burkhart

In fact climate change is already taking an increasing economic toll – from extreme weather events, ocean acidification, desertification in some areas, etc. These costs could increase rapidly if certain tipping points are reached.

But, yes, the larger immediate effects are coming from resource depletion, especially the peaking of conventional oil in 2006. Unconventional oil, like tar sands and fracked oil, is much more expensive, hence produces less wealth, less economic growth. Even much of the newer conventional oil is less productive, as it is often harder to
find or requires tertiary methods of recovery. Similar dynamics apply to coal, natural gas, and many other resources, except that depletion may not be as far advanced as for oil. Economic growth has slowed dramatically even in China, despite their phony growth numbers, and I expect increasing political turmoil there, too, over the next decade or two.

When an imperial economy can longer expand easily, all of Peter’s dynamics come into play with greater force, not just the elite competition, but the increasing exploitation of the common people in order to maintain elite expansion. The latter has been going on since Reagan in the form of escalating economic inequality. = popular immiseration.

Paolo Ghirri

“current problems have nothing to do with anything ecological or resource constraints.”

yes they have: for a pre industrial civilization what is vital is energy surplus, energy surplus that came from agriculture production. so as an example 18 have to work to produce food and 2 can live as soldier, priest and so on. for a industrial civilization energy surplus came from oil. from 1973 to 2016 the energy surplus pro-capita is falling: in a developed country the pro capita surplus now is 75% lower than in 1973. the gap is covered with debt. so in the short run we have: 1) energy price escalation (in real term the 2016 average oil price is the double of 2000) 2) agricultural stress: more frequent spike in food price, combined with food shortfall in the most vulnerable country (arab spring: food price in 2011 are 229% higher than the 2000-2004 average) 3) energy sprawl: investment in energy infrascructure will absorb rising proportion 4) economic stagnation: fail to recover from setbacks as robustly as it has in the past 5) inflation
with the single exception of inflation (but if we check only necessary to live item i’m not so sure) all of the above features has already become firnly established in recent years, wich underlines the point that energy-surplus economy has reached its tipping point

Terry Lowman

The reason the elites cooperate is to get a leg up in the competition. It recently occurred to me that the Forbes 400 list of America’s wealthiest families gives people a rank, a competitor. Without the list, one might be complacent with a mere $3 billion, but knowing others have tens of billions, makes you a “just ran”. Better tune up your capitalist machine so you can outshine everyone else, right?


True but people who cannot be the king of general things will be happy to be known as the king of their specialism.

The more specialisms that exist for people to get to the top of the more stable a society will be.


you could say that the king of the military is the king of kings but in the age of nuclear buttons it’s simply boring. you can’t blow anything up without getting blown up yourself. you can use non-nuclear military power but non-nuclear power in the age we are living in only wins you the war, it doesn’t win you the war and the peace. to win the peace today you need to be king of something other than the military.


Pseudoerasmus, good arguments. The consolidation of money, as well as markets, is very large right now and it does seem like that would take coordination of an ownership class or at least similar lines of thinking among those elites. But, are we talking about a different set of elites? There may be different populations of elites: capitalist and political. Personally, I think the proxies Peter use describe a political elite population rather than a capitalist elite population. The two combine for many, but there may be distinct capitalist and political populations with each having distinct behavior patterns. The worrisome insight for me is that it’s the political elites that end up bringing us to our knees.


“Personally, I think the proxies Peter use describe a political elite population rather than a capitalist elite population.

Political elites are the proxies PT uses as evidence for his theory, but as he himself says, “American power holders are wealth holders”. And I believe the definition I have effectively used here, “owners of capital”, is consistent with his concept of elites or magnates in Secular Cycles — a book I admire tremendously.

Note also that PT uses the Great Merger Movement in US history (1895-1905) as evidence of the beginnings of elite cooperation. Well, another wave of capital concentration has existed now for decades, since the 1980s.

Rich Howard

Political elites may be more likely to be rich, but the rich is a larger population with only a fraction politically aspirant. PT’S model relates political aspirants to political breakdown. And because it works so well, in so many cases, it suggests there is a more universal social process at work than rich/poor, unemployment rates, too many weapons, resource depletion etc.


I like the theory but isn’t there more to the story. On one side you have elite aspirant overproduction. On the other side, you have increasing concentration of power — the iron law of oligarchy (in the sense of this wikipedia link:

Your average Congressman is not as powerful today as he was 100 years ago. Cabinet members used to do something of substance and now act more like front men, while policy making is centralized in the White House.

You have more and more aspirants for fewer and fewer positions of substance. That ramps up intensity of competition even more than just over-production of JDs and MBAs.

Plus the barriers to entry for competition has lowered too. Now celebrities fight with JDs for political positions. Rap stars compete with MBAs for business tycoon success.

At all levels of society, you have greater and greater competition for fewer and fewer rewards. Hyper-competition all around. Now perhaps the competition at the gateway to the elite is particularly important because elites are important, and failure to get in makes them the aspirants powerful disgruntled people, but I think the mechanism is more than just over-production of JDs and MBAs.

I think it might have started as a well intentioned project to increase the quality of our elites by introducing competition and lowering barriers to entry. And at the the same time, increasing the rewards to winners (incentivizing max effort). Result though is brutal intra-elite fighting. Particularly in times of overall lowered growth.

Ross Hartshorn

One point I haven’t seen discussed much is that the number of “powerful” positions is fixed, by law, but not unchangeable. For example, in the 19th century it was arguably more important to be a city councilman or state legislator than a Congressmen, because more actual decisions were being made at the city and state level and the percentage of the economy under the control of the federal government was smaller. If there is less federal largesse to distribute, then there is less power in helping to decide how it is distributed. It is somewhat analogous to why being a U.S. Senator now is more important than being a U.N. functionary; the United Nations may represent a larger domain, but it has a lot less control over that domain than a national government.

Thus, one would expect that the more centralized control of a region is, the more intra-elite competition there will be, because there are fewer positions which really matter. A modern example of this might be that the transfer of power from national to European Union administration would result in more intra-elite competition. On the other hand, devolving power back down to a lower level would result in more positions that have some power, and less competition for each.


That’s exactly what I was getting at too, Ross. The number of good positions available depends on the power gradient of the society. How much power is centralized vs distributed. The whole Iron Law of Oligarchy developed in recognition that over time, power tends to centralize, so it’s not fixed by law and unchangeable for all time. It’s not so much inequality between ordinary people and the elite, but among elites.

Plus it ossifies, in that these enhanced elite positions are then passed out patrilineally, which results in fewer actual positions being open to aspirants.

The net result is heightened competition for entry and promotion within the elite, with more and more of the victories happening by methods outside the norm, e.g. dirty tricks, patronage, fake news etc.

This probably happens in all societies, but growth (creating more opportunities), wars (resetting the table), inefficiency (placating the failed aspirants with consolation prizes) keep internal collapse at bay. It’s when you have a dynamic of High Inequality, Low Growth, High Efficiency / Lean, No Wars that Elite Competition starts getting out of hand.

(I say this despite hating wars, but you can’t argue with their effect on resetting the table. Hate bribes/corruption too, but things like congressional pork barrels kept congressman feeling important and in-line. Efficiency is also a self evident good, but that means no consolation prizes for failure. Growth may eventually run into limits due to carrying capacity of ecosystem….).

To me, it resembles a game of musical chairs with too few chairs, and when the music is playing much too fast. As Chuck Prince famously said in the Global Financial Crisis: “As long as the music is playing, you’ve got to get up and dance.” Whether or not dancing is destructive, elites have to keep dancing to keep their chair.

Ross Hartshorn

I also hate wars, but I am reminded of Mancur Olson’s theory that nations recovering from a major disaster or a major military defeat usually have above-average growth for a few decades. The idea is that when, as with the South in the U.S. after the Civil War or with Germany and Japan after WWII, the elite in society have suffered a setback so severe that their hold on society is disrupted, there will be a period during which they are less able to set government policy in their favor rather than the collective welfare.

SDT would have a somewhat different explanation of this. I agree with you that rapid growth would be another way to reduce the intra-elite competition; it seems the most likely explanation for the “missing” peak in non-governmental violence in the U.S. in the 1820’s that Peter Turchin pointed out earlier.

Ross Hartshorn

This idea is kind of half-formed, but I’ll put it out there. It seems to me that one of the most important factors in intra-elite competition, is the degree of skill of the frustrated aspirants. If there are lots of people who want to be elite but can’t crack the system to get in, that may not be a problem if those frustrated aspirants aren’t particularly good at organization, motivation, leadership, etc.

If, on the other hand, the frustrated aspirants are nearly as good at this sort of thing as those actually in power, and especially if they are better at it than the incumbents (who somehow through tradition or family connections or what-have-you remain on top), then you have a much better chance of the frustrated aspirants being able to kick up trouble.

Of course, part of being good at leadership is getting the opportunity to practice, and a post-secondary education almost always includes some practice at a more professional set of social skills. But if the people getting spots in power remain better at political organization than the people who don’t, it is less likely to result in disruption, I think. It seems that trouble would come when the ruling elite is either not especially good at leading (e.g. they inherited their position or bought their way in with somebody else’s money), or they were good at leading in a previous time, and changes in society or technology have changed what skills are necessary for leadership.

In all these cases, I think “good at leadership” would be a relative term, which is to say the current elite relative to the frustrated aspirants. How you could measure such skill, of course, is the key question about which I have as of yet nothing to say (I did say the idea was half-formed).

steven t johnson

Although intraelite competition and interelite competition are conceptually distinct, is that true in practice? Is Carlos Slim an intraelite competitor with Jeff Bezos, in the form of rivalry between the New York Times and the Washington Post? If this is interelite competition, how does structural-demographic theory address the issues of how external factors impinge on the cycle? (I’m a little shaky on how interior and exterior are defined in the first place. As for example, was there a cycle for Burgundy?)


//The supply of power positions in a society is relatively, or even absolutely, inelastic. For example, there are only 435 U.S. Representatives, 100 Senators, and one President. //

This is not quite true. The supply of power positions can be elastic to a point.

How about the growth in number of CEOs and NGOs and the heads of INGOs over the last 50 years? So-called non-state actors have become powerful as they influence the law-making processes in a variety of ways.

These big chiefs are positions of power and influence. In many cases, they call the shots and Presidents and Prime Ministers are only the PR guys.

The US President is not the most powerful person in the world. He doesn’t have the highest security clearance in the United States. He is not allowed to know everything.

The idea the US President is the most powerful man is a claim based on a theory of how the US political system works in idealised sense, and on simple US nationalism.

The fact that the supply of power positions is elastic – that there has been a flouresence of alternative power structures to the state hierarchy – suggests that wealth can to a degree put off or delay elite competition.

It is only when the rug is pulled from under the alternative prestigious hierarchies and the state tries to dominate all on its own – that is when problems will begin. Keep the funding going, maintain non-state avenues for prestige and create even more, the fluoresence will continue.


interested readers might like to read my report for Cliodynamics: Why Has the Number of International Non-Governmental Organizations Exploded since 1960?


I agree, also see comment #2

Nikhil ns

A point made in arthashastra, that fight among princes is more dangerous than fight among commoners. However, I wud like to ask what predictions are u unable to do. There is no real knowledge which doesnt admit what its limitations are, or admits inability to explain something. Even in physics, where humans have gained incredible knowledge, there is much to know. Also, on issue of religion, could one argue that but for christianity & islam world wud have devekped faster as information in math/science wud have gathered pace, exchanged between different lands easily.Thank you.

Stephen Morris

What is extraordinary about this article is that the author seems to accept – as a matter of Faith – the necessity of Elite rule. It’s as if Alexander Hamilton had risen from the dead and started write blogs.

We have the dogmatic assertions such as:

“Maintaining such norms is the job for the elites themselves.” What if the 99% disagree with those norms? We have the seen the damage inflicted by “Elite Norms” in such things as the Washington Consensus or the Neo-Liberal Consensus (which, like Elite norms before them are really nothing more than a justification for Elite privilege).

Or this:

“Intra-elite competition, thus, has a nonlinear effect on social function: moderate levels are GOOD [emphasis added], excessive levels are BAD [emphasis added].” That’s a matter of opinion. If high levels of intra-elite competition have the effect of overturning the Elite in favour of a more democratic power structure, who is calling that “BAD”?

This is especially relevant to the non-democratic system of purely “elective” government which operates in most polities. We have seen in the work of Gilens and Page that elective government governs:

a) typically for the benefit of (at most) the wealthiest 10% of the population; and

b) occasionally for the benefit of certain well-organised pressure groups whose preferences do not mirror those of the majority of citizens.

Moreover, as Economics Nobel laureate James Buchanan explained, such systems tend to “adversely select” the most aggressively narcissistic, machiavellian (and possibly psychopathic) individuals to positions of power:

“[S]uppose that a monopoly right is to be auctioned; whom will we predict to be the highest bidder? Surely we can presume that the person who intends to exploit the monopoly power most fully, the one for whom the expected profit is highest, will be among the highest bidders for the franchise. In the same way, positions of political power will tend to attract those persons who place higher values on the possession of such power. These persons will tend to be the highest bidders in the allocation of political offices. . . . Is there any presumption that political rent seeking will ultimately allocate offices to the ‘best’ persons? Is there not the overwhelming presumption that offices will be secured by those who value power most highly and who seek to use such power of discretion in the furtherance of their personal projects, be these moral or otherwise? Genuine public-interest motivations may exist and may even be widespread, but are these motivations sufficiently passionate to stimulate people to fight for political office, to compete with those whose passions include the desire to wield power over others?” (James Buchanan and Geoffrey Brennan, “The Reason of Rules”, Cambridge University Press, 1985, p64)

The solution to adverse selection of aggressively narcissistic elitists is to eliminate the Monopoly on Power through Democracy: genuine Democracy with the right of recall, veto, initiative and referendum. As Buchanan wrote elsewhere:

“In sum, the effects of direct democracy add-ons to existing decision rules surely work toward reducing the range and scope for politicization, a result supported by classical liberals.”

This brings us to the final paragraph in which Peter Turchin can see nothing but bad coming from a period of intra-elite competition (“a massive outbreak of political violence, often ending in a state collapse, a revolution, or a civil war”).

In fact, the last outbreak of Populism in the United States had a much more beneficial and peaceful legacy. That was the introduction of genuine Democracy (at the state level) into about half of the US States.

(Admittedly, those partially democratic States must still operate under the savagely anti-democratic provisions of the Federal constitution and its Supreme Court which – in the name of the “Rule of Law” – insists that Money is Speech and overturns attempts to regulate the role of money in the democratic process. But that’s hardly a shortcoming of Democracy.)

We know from work such as that of Bower et al (“Enraged or Engaged? Preferences for Direct Citizen Participation in Affluent Democracies”, 2007) that:

a) in almost all countries a clear majority of respondents agree or strongly agree with the statement “Thinking about politics in [COUNTRY] . . . . Referendums are a good way to decide important political questions”;

b) in countries where there is no outright majority support, a strong plurality of respondents agree or strongly agree (with some having no view); and

c) support is STRONGEST in that country (Switzerland) where people have the MOST experience of such decision-making.

We know from the historical record that in the few cases where citizens HAVE been given a free choice in their system of government they almost invariably vote for genuine Democracy with the right of recall, veto, initiative and referendum.

Most importantly, we know that where citizens DO enjoy truly democratic rights they NEVER vote to repeal them, even though it’s a straightforward process to initiate a referendum for that purpose. (And indeed in some jurisdictions the attempt has been made . . . and defeated at the ballot box!)

Unlike the system of elitist elective government, genuine Democracy demonstrates the ongoing consent of the citizens being governed.

So the question for the Elites to answer must surely be this:

“When were they granted their ‘Charter from Heaven’ authorising them to deny the citizens the right to choose the form of government THEY prefer for THEIR country or State?”

Dick Burkhart

Even direct democracy is not a cure-all. Here in Washington State, our initiative and referendum process has been corrupted at times by big money interests: First put together a sophisticated campaign around some catch phrases that will have popular support on a topic where the opposition, even if widespread, is likely to be diffuse. Then sneak in some coded language that privileges a wealthy special interest. Then use paid signature gatherers. Then assemble a massive advertising campaign, one that will outspend the likely opposition, maybe even by 10 to 1.

Certain people get very good at this and quickly learn to sell their services to the highest bidder. The current master of such campaign here is a guy named Tim Eyman, and he has been quite successful. But some companies, like Costco, have done the same thing all by themselves.

Moral: You need to get “money out of politics” in all ways, and it’s a never ending battle until you’ve eliminated concentrated wealth and power itself.

Stephen Morris

The question put earlier was not:

“Do you personally prefer genuine Democracy?”

It was:

“When were [the Elites] granted their ‘Charter from Heaven’ authorising them to deny the citizens the right to choose the form of government THEY prefer for THEIR country or State?”

What we observe in Washington State and other polities which enjoy some form of genuine Democracy is that the citizens never vote to abolish it, even though it is a straightforward process to call a referendum for that purpose. In some cases (for example, three times in California) there have been attempts to do just that, but they have always been defeated at the ballot box or withdrawn in the face of imminent defeat.

Whether or not one personally prefers genuine Democracy is irrelevant unless one can show that one’s preferences on the matter ought to be privileged over those of other people, people who apparently prefer genuine Democracy ON BALANCE over non-democratic forms of government.

In general, any attempt to demonstrate that one’s own preferences ought to be privileged will reduce either to:

a) another statement of preference, which begs the question; or

b) some identifiable logical fallacy.

In relation to the specific claims made in the preceding comment, it should be noted (as mentioned earlier) that the problem of money in politics is not a problem of Democracy but of the lack of Democracy. Several democratic States have sought to regulate it (especially paid petition collection) but their attempts have been nullified by the ferociously anti-democratic US Supreme Court which – in the name of the “Rule of Law” – has determined that Money Is Speech and that attempts to restrict money are an infringement on free speech (Meyers v Grant, 1988. See:

In any event there is some doubt over whether money has any effect on referendum outcomes and what that effect might be. Some studies suggest that money works to get an initiative onto the ballot paper in the first place but does little to guarantee its final approval.

Other studies suggest that money can have an effect on initiative outcomes, but that comes from changing voter attitudes rather than altering voter turnout. This raises two important matters of principle:

a) if at the time of voting, voters actually do prefer a particular outcome (even as a result of intensive campaigning) on what grounds can it be claimed that their preferences are “wrong”? That requires someone to second-guess what is “right” and “wrong” and to establish that their own preferences are “privileged”; and

b) is that outcome worse than not allowing voters the possibility of initiatives at all? It is all very well to point out possible defects in genuine Democracy, but if the alternative is elective government – also corrupted by money politics – is that any better?

Finally, there are other models of initiation which overcome some of these issues. For example, the Swiss Constitution allows not only popular initiative but also initiative by one third of the cantons. One might consider the possibility of:

a) initiation by local authorities at the State level;

b) initiation by, say, one third of States at the federal level. At present the US Constitution provides that TWO-thirds of States may call for a Constitutional Convention but it is not clear what form that convention would take, whether it would be effective in getting the specific grievance handled, and how any ratification might take place; and

c) initiation by one House of the Legislature to resolve a deadlock by reference to the citizens. This has been used by governors at the State level but could conceivably be applied more widely. It is not widely known that such a scheme was proposed by the British House of Lords as an alternative to the Parliament Act of 1911 (the effect of which was to create Britain’s now notorious “elective dictatorship”). The proposal was never put to the voters. Indeed, it was not even publicised at the time. One might wonder how differently Britain might have developed had it introduced a form of Democracy in 1911?


You know, Switzerland is an interesting case for Peter (and others) to tackle.. They have had wars between cantons (earlier in its history), but for the most part, they have been able to remain stable for centuries. Is it because it is decentralized? Small? The referendums? All of the above?

I didn’t get the sense that they are terribly egalitarian. . . .

Stephen Morris

An interesting aspect of Swiss Democracy is its effect on elective government.

The Swiss Federal Cabinet is bipartisan. Indeed it is pan-partisan comprising ( seven members drawn from the four major parties across the political spectrum. It is not uncommon for a Swiss Minister to be responsible for implementing policies at odds with his or her own party’s platform.

The plausible reason for this is that Switzerland is a genuine Democracy. The citizens have the ultimate say in what goes on through the mechanism of initiative and referendum. As a result, Swiss Democracy does not attract the aggressively narcissistic megalomaniacs commonly seen in corrupt elective systems.

There is no concept of “Government” and “Opposition” battling to see who will rule.

There is no infantile obsession with “Leaders”. Democratic politicians are not there is “lead” but to represent.

On the matter of decentralisation, it is not only a question of whether Democracy works best in a decentralised polity, but whether centralisation itself relies on NON-democratic government.

The evidence for popular attitudes towards centralisation/self-determination is thin because people are so very rarely given an opportunity to express a view on the matter in any binding form.

We do, however, have some evidence.

In Australia where the Constitution requires a referendum to approve any changes, we know that there have been 32 attempts to increase the power of the central government. Of these, only 2 were ever approved (the Commonwealth Social Security referendum of 1946 and the transfer to the Commonwealth of power to make law “with respect to . . . the aboriginal race in any State” in 1967). Of the remaining 12 referendums (dealing with ancillary matters such as retirement of judges and agreements over State debts) half were approved, suggesting that popular opposition was toward centralisation of power rather than to constitutional change per se. Similar rates of approval are seen in State referendums. (As is well known, the Australian central government has had its powers increased anyway through the High Court – appointed by the central government – reading broader and broader meanings into the existing language of the document.)

In Europe we saw the “European Constitution” – a centralising proposal – voted down in the very heartland of the EU, France and the Netherlands. The politicians had to suspend any further popular voting and ram it through without popular consent (other than in Ireland which constitutionally requires a popular referendum on all such treaties) in the form of the Lisbon Treaty. Even in Ireland it was defeated on the first round and approved only after special pro-Irish provisions had been appended. Other European Peoples were given not such special treatment.

In Europe also, the only two countries given a say on the centralisation of monetary policy (Denmark and Sweden) both voted NO, despite both wing of the Establishment urging them to do so.

In Britain, in the only case of people being allowed a vote on whether to withdraw from the EU (which had vastly expanded its powers since the original referendum) we saw a YES vote.

Earlier in Britain, we saw the Scottish people vote to remain part of the United Kingdom but only after being given promises of extensive devolution of power. The Scots are now weighing up the relative merits of a remote government in London and a remote government in Brussels.

In Switzerland in 1979 we saw the people of Jura vote to leave Canton Berne.

And of course there are obvious cases such as East Timor which in 1999 voted to leave Indonesia.

In contrast, popular votes in favour of centralisation typically involve the delegation of very limited powers. Once a central government has been formed, however, the “Iron Law of Megalomania” takes hold and it attracts to itself the machiavellian narcissists who then campaign to expand their powers without popular consent.

The evidence suggest that communities do prefer to govern themselves, that centralisation is favoured by aggressively narcissistic megalomaniacs, and that genuine Democracy might prevent that.

On topic of egalitarianism in Switzerland, it is plausible that people are less concerned by large inequalities of WEALTH if they can be assured that those inequalities will not translate into inequalities of POWER. Genuine Democracy may provide such assurance.

Stephen Morris

The question put earlier was not:

“Why do you personally prefer loathe Democracy?”

The question put earlier was:

“When were [the Elites] granted their ‘Charter from Heaven’ authorising them to deny the citizens the right to choose the form of government THEY prefer for THEIR country or State?”

Peter Turchin’s depiction of imagined chaos goes a long way to explaining the squalid camps one finds in France, Germany, Italy and Austria, filled Swiss refugees escaping the horror and privatation of life under Democracy(!)

More seriously, Peter Turchin’s response demonstrates the essential problem facing fundamentalist Elitists: they are faced with trying to defend the indefensible.

And so they resort to hypothetical evidence and logical fallacies.

Meanwhile, back in the real world of evidence and logic, we find that:

a) wherever people are given a free choice, they almost invariably choose Democracy to the greatest extent made available to them; and

b) where people enjoy Democracy, they NEVER vote to abolish it, even though it is a straightforward process to initiate a a referendum for that purpose.

So perhaps one day if Peter Turchin finds the courage, he might provide an answer to the question that was put:

“When were [the Elites] granted their ‘Charter from Heaven’ authorising them to deny the citizens the right to choose the form of government THEY prefer for THEIR country or State?”


The supply of status positions is not fixed and is in fact quite elastic when there is economic growth which is highly concentrated at the top. Status aspirants have many avenues of status generation. A good approximation is that the number of positions at the top is proportionate to economic growth. On Twitter, Peter Turchin mentioned “There are only 500 CEO positions in Fortune 500 companies”. But there’s also the Fortune 1000 — the market capitalisation of the 1000th company is $17 billion! The average capitalisation of the Wilshire 5000 and Russell 2000 is $530 million and $253 million, respectively. Equivalent numbers in real terms were *much* smaller 20 years ago.

In other words, economic growth keeps increasing the value of ‘status’ positions in society.

But Peter Turchin would have us believe that the subjective value of status positions for the status holders is purely ordinal — what they care about is mostly getting ahead of others, so elite aspirants are always competing to get to the Nth place when they are at (N-1)th, and this is an important cause of sociopolitical instability.

Unfortunately, lawyers’ salaries and indices of congressional polarisation are very meager evidence for such claims.

Peter Turchin’s model of elites is based on agrarian societies where social status is conferred by the fixed number of administrative positions in the state and in the size of the parcels of land allotted to them. This model is outdated in modern industrial societies with economic growth.

PT says my view of status is too monetary. Well, not really. Anyone who is rich in the USA buys status through fame, social status, and conscpicuous consumption. Who has more status in US society? Justin Bieber with 30 million Twitter followers, or the senior senator from Nebraska?


This is the correct link for the market capitalisation of the 1000th company

And I meant the *median* market capitalisation not average for the Wilshire 5000 and Russell 2000


That’s why i say it’s not just the number of elites, it’s their relative security. There are a lot more CEO’s but the average length of CEO tenure has dropped.

Elites are feeling less secure than ever. That’s before you count in populist resistance. They aren’t looking to be N-1 as much as they’re worried about becoming N+1… or worse yet, falling out of the virtuous circle altogether.

Plus discretionary power has been reduced too for elites. CEOs have to personally attest to truth of SEC filings, follow compliance and ethics codes. More watchdogs, more audits, more open mics or candid videos captured on social media.

Elites are paid well (better than ever), but it’s not the secure power position elites used to have. More people want to be elite, or feel entitled to it, but truly # of truly elite positions have been reduced.

Paolo Ghirri

“with economic growth” Between 2000 and 2015, and expressed at constant 2015 dollar values, global real GDP expanded by $27 trillion – but this came at the expense of $87 trillion in additional indebtedness (a number which excludes the inter-bank or “financial” sector). This meant that, in inflation-adjusted terms, each growth dollar cost $3.25 in net new deb. is economy that growth or debt that give illusion of growth?

Greetings! I know this is kind of off topic but I was wondering which blog
platform are you using for this site? I’m getting fed up of
Wordpress because I’ve had problems with hackers
and I’m looking at options for another platform.
I would be awesome if you could point me
in the direction of a good platform.


Prof Turchin, is there any data on the Supply of Elite Positions in Historic Societies?

It doesn’t feel instinctively right that it’s inelastic, but perhaps there’s really the case. It feels slightly more likely to be right to say that it’s capped somehow (inelastic as to upside, more elastic as to downside).

But it seems like the sort of thing you should be able to answer with a History Database. Has there been any attempts to measure this?

[…] mistaken about elite competition in modern societies” writes Pseudoerasmus in a comment on my previous post. So let’s strip the argument down to its most basic form. In my post I explicitly use the […]

[…] 10 Episodes of Television Every Catholic Should Watch – EpicPew. CAN THE REPUBLICAN PARTY AVOID THE FATE OF THE WHIGS? Gun Control Charts! The Eponymous Flower: The War Against Reality. THE HERETICAL (OR CHRISTIAN-INFLUENCED) VIEW OF MODERN MARRIAGE. Marriage, Procreation and Historical Amnesia. ‘Hyperlocal’ Web Sites Deliver News Without Newspapers. 52 Of The World's Most Widespread Myths And Misconceptions, Debunked. Crunchy Granola Catholics. Peter Turchin Intra-Elite Competition: A Key Concept for Understanding the Dynamics of Complex Socie…. […]


Deficiencies in the concept of elite competition
Let’s start with the definition of elite: “small proportion of the population that concentrates power in their hands”
His theory lacks an aspect that must be fundamental before even proceeding in a discussion on the “dynamics” of the elites and is that it is not able to explain in a satisfactory way the origin of the so-called “elites”. According to its definition it seems that the elites are rather the manifestation of a particular phenomenon that is “concentration of power”; A phenomenon that manifests itself socially in the form of the so-called “elite”, which hereafter I call the ruling class (I think it is a terminology in which we can all agree).
But if we assume that the dominant classes are only a manifestation of the phenomenon of the concentration of power, our attention must first be fixed in that aspect so we try to break it down into its fundamental parts
. Apparently the concept of power gives to understand the concept of dominion (some will have other words in mind but as surely they closely resemble the concept of domain I think that it suffices to refer us to this one) and we do not refer to any type of domain but to a domain Of social nature, a social domain. We will now say that this social domain manifests itself in the form of economic and political dominion, I think we will agree on this point.
Now let us collect the fruits of these arguments. We have a different and more precise definition, which in no way invalidates the original, and we say: The ruling class is that small proportion of the population that concentrates economic and political dominion in their hands. I believe that we will agree that economic dominance is nothing but greater possession of capital and that political dominance is but a major influence on a state structure (the word “state” is used in a modern sense).
Now we have: the ruling class is that small proportion of the population that concentrates the greatest possession of capital and the greatest influence within a state structure in their hands. The last part of “… in your hands” is understood by what we can eliminate it and we have the following:
The ruling class is that small proportion of the population that concentrates the greatest possession of capital and the greatest influence on a state structure.
Now the possession of capital depends on its production or of the association with someone who produces capital. And it is revealed to us that the ruling class, apart from having influence in a state structure, needs to produce capital or be associated with someone who produces capital directly or indirectly.
Thanks to this we see clearly that competition between elites is a competition for economic benefits and influence. Obviously the economic aspect is more significant than the aspect of influence. It follows that a fall in economic profits, ie a fall in capital production (a crisis), would directly or indirectly exacerbate the competition for greater economic benefits, that is, increase the number of aspirants to elitist . The competition of elites is not the cause of the crisis is one of the consequences of the crisis.


I must make a small correction in my analysis. By capital I wanted to let you understand profit, so the use of that term in this argument is actually inappropriate because I wanted to use the word capital in a Marxist sense.

republic day quotes

Amazing! This blog looks exactly like my old one!
It’s on a totally different subject but it has pretty much the same layout and design. Outstanding choice of colors!

gadget scanner

I was curious if you ever considered changing the structure of your blog?
Its very well written; I love what youve got to say.
But maybe you could a little more in the way of
content so people could connect with it better.

Youve got an awful lot of text for only having one or 2 images.
Maybe you could space it out better?


Hello Dr Turchin, I was wondering if you are familiar with Richard Lachmann’s “elite conflict theory”. It is a verbal theory, but one that he has successfully used to explain fiscal crises, hegemonic cycles, and the rise of modern capitalist economies. What do you think about it?

Ncert books Solution

Excellent goods from you, man. I have understand your
stuff previous to and you’re just extremely fantastic. I actually like what
you’ve acquired here, really like what you’re saying and the
way in which you say it. You make it entertaining and
you still care for to keep it smart. I cant wait to read far more from you.?

[…] Source: Peter Turchin Intra-Elite Competition: A Key Concept for Understanding the Dynamics of Complex Socie… […]

Greetings! I’ve been reading your weblog for a while now and finally got the bravery
to go ahead and give you a shout out from Atascocita Texas!
Just wanted to tell you keep up the excellent work!

[…] mistake to assume that one faction or the other will prevail. At least, that’s the implication of a recent essay by Peter Turchin, a Russian-American ecologist specializing in the study of cultural evolution. Without specific […]

[…] mistake to assume that one faction or the other will prevail. At least, that’s the implication of a recent essay by Peter Turchin, a Russian-American ecologist specializing in the study of cultural evolution. Without specific […]

[…] mistake to assume that one faction or the other will prevail. At least, that’s the implication of a recent essay by Peter Turchin, a Russian-American ecologist specializing in the study of cultural evolution. Without specific […]

Shaun Bartone

I wonder if any of the commentators here have considered that the cabal now in power in the US (not elsewhere) are not in power to “take power” except for a temporary period. They don’t want to run the federal government, they want to destroy it, except for the police state and the military. They want to eliminate the EPA, vacate the State Dept and many other Depts, except for a few high-placed cronies, wipe all financial, labour, consumer and environmental regulations off the books; eliminate or reduce to a bare minimum federal health insurance, medicaid, medicare and Social Security, crush public education, privatize everything they can sell, and so on. They are not in power to “govern” but to destroy government. This is all being done with a fairly unified agenda: to free “the market” from any restrictions whatsoever, so that they—global elites —can make as much money as possible. It’s a cabal of global corporations, militarists, Christian sovereign white supremacists, fossil fuel giants and bankers, and I think there’s a high degree of cooperation for the agenda. The revolution is the cabal run by Trump/Bannon who are more extreme and ideological than any previous faction, who have no tolerance for compromise. They have an apocalyptic vision of grinding it all down to a bare minimum police state.

[…] Peter Turchin: Intra-Elite Competition: A Key Concept for Understanding the Dynamics of Complex Societies […]

  1. Home
  2. /
  3. Cliodynamica
  4. /
  5. Regular Posts
  6. /
  7. Intra-Elite Competition: A Key...

© Peter Turchin 2023 All rights reserved

Privacy Policy