The Stationary Bandit Model
In the minds of many, the only alternative to
democracy is dictatorship.
Tullock’s Question - “What’s
so bad about dictatorship anyway?” – highlights the fact that
public choice economists have spent little energy analyzing history’s typical form of government.
Simplest approach: Dictatorship is equivalent to
democracy with a single - and perfectly decisive - voter.
All of the usual rules about democracy that
hinge on low probability of decisiveness reverse:
Thus, we should expect dictators to be highly
self-interested, but more interested in rationally assessing policies' actual
Will this lead to bad consequences for the
According to McGuire and Olson’s “stationary bandit
model,” not necessarily. As
long as the dictator knows that he will be around for a long time, it is in his
rational self-interest to encourage/allow economic development – to take
a smaller slice of the pie in order to make it grow faster.
Alternate perspective: Stationary bandits go to
the maximum of the long-run Laffer
curve instead of the short-run or instantaneous Laffer curve.
Remember the Tiebout model? It is basically a model of dictatorship
constrained by mobile capital and labor, and under standard assumptions it
yields perfectly competitive results.
If the rulers of Tiebout governments were really
dictators, then my arguments about non-profit competition would no longer
In the real world, dictators often respond to
the mobility of capital and labor by trying to make them less mobile. The Berlin Wall is the most notorious
– but not the most horrific – example.
However, dictators do treat mobile resources
better. East Germany rarely forced
tourists to become citizens, and Communist governments rarely defaulted on
their sovereign debt.
Many dictators go further by making war to get
more resources under their control.
Why grow when you can conquer?
Another reason for dictators to stifle growth:
Growth leads to contact with the outside world and/or free thought, which tends
to undermine the dictator’s authority.
Constrained Dictatorship and the Paradox of
Very few dictatorships actually fit the
“one decisive voter” model, though modern totalitarian regimes
– like Stalin’s USSR, the Kims’ North Korea, and the last
years of Hitler’s Germany – come close.
Almost all dictatorships throughout history have
instead been “authoritarian.”
The dictator has a lot of say, but at least de facto, so do many other
actors. The dictator ignores them
at his own risk; if he goes over the line, he risks a coup.
Most people add that at some point, an abusive
dictator would provoke popular resistance.
Mises argues that this threat is so strong that
dictatorships follow exactly the same policies that democracies would
have! I call this his
“Democracy-Dictatorship Equivalence Theorem.”
Tullock, in contrast, argues that collective
action problems make popular revolutions virtually impossible.
Most political observers believe in the
existence of revolutions, so for them Tullock’s argument creates a
“paradox of revolution” – revolutions seem impossible in
theory, yet they occur.
For Tullock, however, “popular”
revolutions are thinly disguised battles between rival elites. The competing sides solve their
collective action problems with selective incentives – better ration
cards, promises of post-revolutionary jobs, etc.
Trotsky’s on Tullock’s side: "An army cannot be built without
reprisals. Masses of men cannot be led to death unless the army command has the
death penalty in its arsenal. So
long as those malicious tailless apes that are so proud of their technical
achievements - the animals that we call men - will build armies and wage wars,
the command will always be obliged to place the soldiers between the possible
death in the front and the probable one in the rear."
Watered down version of Tullock: Revolutionary
movements require true believers to get off the ground, but further growth
requires selective incentives.
The Sociopathic Bandit Model?
A major complication for economic models of
dictatorship: Being dictator effectively makes someone extraordinarily
wealthy. The resources of an entire
nation are theirs to command.
Due to their extreme wealth, they may consume a
lot of altruism, expressive considerations, and/or irrationality despite their high price.
Hence we see all sorts of strange behavior from
Mass murder of seemingly useful subjects.
Awe-inspiring parades, monuments, palaces, etc.
Bizarre social experiments.
And… voluntary reduction to figurehead
Modern dictators rarely accept figurehead status,
but it happened all over 19th-century Europe when traditional monarchs
allowed and even urged a move to “constitutional monarchy.”
Tullock’s explanation: The selective
pressure for power-hunger is much weaker in hereditary dictatorships.
On balance, then, it is hard to make definite
statements about the selfishness, instrumentalism, or rationality of
dictatorial versus democratic policy.
It’s got to be studied empirically.
The most convincing claim economic theory has to
make about dictatorship: It's a big gamble. Everything depends on the idiosyncrasies
of the Leader. This makes sense in
theory, and works empirically:
Interesting finding #1: Average growth of
dictatorships and democracies is the same, but dictatorships have more
dispersion. Graph from Almeida and
Interesting finding #2: When a dictator accidentally dies, growth rates
persistently change. When a
democratic leader accidentally dies, in contrast, they don’t.
Political scientists distinguish between
“totalitarian” and “authoritarian” dictatorships. Standard totalitarian checklist courtesy
of Richard Pipes:
official all-embracing ideology
a single party of the elect headed by a
“leader” and dominating the state
the ruling party's control of the means of
communication and the armed forces
central command of the economy
Since #5 is equivalent to socialism in the
traditional sense of the word, many socialists object to this criterion. But it is hard to rebut Trotsky’s
explanation: "In a country where the sole employer is the State,
opposition means death by slow starvation. The old principle: who does not work
shall not eat, has been replaced by a new one: who does not obey shall not
One cheer for democracy: Totalitarianism has
almost never been established democratically. (Semi-convincing counter-example:
Hitler’s Germany). It arises
through civil war (USSR, China, etc.) or conquest (Eastern Europe, North
A few analytical narratives on the rise of
A few analytical narratives on the implosion of
Is Totalitarianism Possible? Economic Calculation Reconsidered
Austrian economists were harsh critics of
totalitarianism before it existed. So
was everyone sensible. The uniquely
Austrian objection was that Characteristic #5 (socialism) is
Why is socialism impossible? Mises’ original argument:
Economic calculation (comparing the cost of
different ways of doing the same thing) requires prices.
Prices require some kind of market (not
Under socialism, there is no market, therefore
no prices, therefore no calculation.
Socialism is impossible.
For Mises, “impossible” means total social collapse! “[T]he
attempt to reform the world socialistically might destroy civilization. It would never set up a successful
“Socialism cannot be realized because it is beyond human power to
establish it as a social system.”
socialists replied that market socialism or faster computers would make
socialism possible, but the rejoinders are obvious.
My complaint: The argument is fine until the
conclusion! The lack of economic
calculation makes socialism more difficult,
but difficult is not impossible.
Furthermore, the economic history of socialism
Its biggest disasters – massive famines
where millions died – were
caused by bad incentives, not lack of calculation.
Socialist planners habitually ignored capitalist prices; they
didn’t just preserve socialism by free riding on the price system of the non-socialist
When socialist societies wanted results, they
used strong incentives and got results.
See their secret police, border security, militaries, space programs,
Olympic teams, and nuclear weapons.
Also note: Incentive experiments in Soviet
agriculture showed that it was possible to sharply raise output, but the
experiments were ignored and their initiators punished.
Democratic Transitions: What Happens?
One fact that Mises’ Equivalence Theorem
can explain, and Tullock can’t: When dictatorships peacefully become
democracies, policy usually doesn’t change that much. Examples:
Strong populist back-lash against free-market
policies – and election of ex-Communists (and even unrepentant
Communists) – in the former Soviet bloc.
Chile kept most of Pinochet’s economic
policies after he relinquished power.
Free elections in Palestine did not lead to
However, there are many more plausible
explanations than Mises’ story that dictators are fully constrained by
the threat of revolution.
The stationary bandit model. Stable dictators, like the median voter,
benefit if their country has pro-growth policies.
Of course, this model isn’t very
convincing if you think that both kinds of governments have deeply inefficient
Shared preferences. Especially in long-lasting dictatorships,
the dictator and the median voter come from the same basic political culture
and therefore have similar preferences.
quo bias. To a large extent,
dictatorships successfully brainwash their populations to think that what is,
is a pretty good idea.
result: Alesina and Fuchs-Schündeln
(2005) find that East Germans are markedly more
anti-market than West Germans, even controlling for income. Living under socialist tyranny
doesn’t make people hate socialism.