put on the various sides, and a whole series of peace settlements
achieved: Sri Lanka, Burundi, Southern Sudan, Sierra Leone,
Angola, the Democratic Republic of the Congo, Bosnia, and Ko-
sovo. While this was a splendid achievement, post-conflict situ-
ations are fragile; in the past around 40 percent of them have
reverted to violence within a decade. In total these reversions
account for around half of all the world’s civil wars. So main-
taining the post-conflict peace more effectively than in the past
would be the single most effective way of reducing civil war. Is
democracy the key to peace in these societies? International ap-
proaches to post-conflict situations are still in their infancy: a
new organization, the Peace-Building Commission of the United
Nations, is just finding its feet. The recent record is not entirely
encouraging: here are a few examples.
Take the transitional government of the Democratic Repub-
lic of the Congo. Knowing that they had only three years in power
76
WARS, GUNS, AND VOTES
before facing elections and the possible loss of office, ministers set
about plundering the public purse. But the public purse was pretty
small because tax revenue had withered away: as you will see, low
taxation is part of the strategy of misgovernance. But plunder can
extend beyond tax revenue. One strategy would be to borrow: sad-
dle future citizens with liabilities and run off with the proceeds. Un-
fortunately for the new leaders of the Democratic Republic of the
Congo, this strategy was not feasible: President Mobutu had already
used it to the hilt so that the country was beyond its neck in debt. No
bank was going to lend.
But there was an alternative. The Congo is mineral-rich. Much
of these resources are unexploited because under President Mobutu
it would have been folly for a company to incur the investment nec-
essary to sink a mine. The president was stuck in what economists
call the time-consistency problem: because he could not bind himself
from confiscating investments, no sane company would make them.
But by the time of the transitional government the global boom in
commodity prices had changed the calculus of risk: it was worth
paying a little something for the exploitation rights that the transi-
tional government could legally confer. And so the ministers of the
transitional government of the Democratic Republic of the Congo
mortgaged the future of its citizens as surely as if they had issued
debt, by selling off national assets at bargain prices. A few months
ago I had lunch with one of the shrewd purchasers of these rights: a
good lunch it was too. He became a little upset when I told him that
the rights ought to be renegotiated.
Now take the most remarkable of all the conflict settlements,
the peace in Southern Sudan achieved after many years of violence.
The new government of Southern Sudan inherited an economic
landscape that was virtually lunar: no provision whatsoever of pub-
lic goods. No roads, no schools, no health care: nothing, not even
buildings. The only public good was the security force, the Sudanese
People’s Liberation Army (SPLA), and that had just become redun-
Inside the Cauldron
77
dant. There were, however, massive financial resources. Southern
Sudan was sitting atop a newly opened oil field that straddled the
border with Northern Sudan, and its share of the oil revenues pro-
vided an instant flow of $1.3 billion per year. On top of the oil rev-
enues there was a huge aid inflow: quite appropriately, every agency
wanted to help.
This was an environment par excellence in which priorities and
sequence mattered. After all those years of sacrifice in the cause of
liberation, the inhabitants of Southern Sudan might reasonably have
hoped that their government would think through the critical path
of building an effective state and get on with implementing it. So,
two years on, what has happened? As a senior minister put it to me,
“We’ve lost it.” The most serious error was to devolve the power of
public spending to the commanders of the military units that made
up the SPLA. What did they do with their power? They expanded
their own fighting units, putting their soldiers on the new public
payroll. This alone has exhausted the oil revenue. And, of course,
the government is now stuck. It can only free up the budget for
productive uses by dismissing fighters who have just got themselves
onto the gravy train. What else have ministers done: what about the
aid money? Ministers themselves have decided not to bother liv-
ing in Southern Sudan: they live in Nairobi where the public goods
are better. They commute into the country they are responsible for
governing because the donor agencies insist on holding meetings
there. So what do the meetings reveal about ministerial priorities?
Priority number one is for large, imposing ministerial headquarters
buildings: you can surely picture the designs for the soaring concrete
structures that will be the ministries of this and that.
I tend to think that governments get the private sector they de-
serve. In Southern Sudan there is one huge private investment. It is
a five-star luxury hotel, sitting, like a space hotel, in the middle of
nowhere. Because of the absence of public goods, there isn’t even
a road that leads to it. Who are the intended clients for the hotel?
78
WARS, GUNS, AND VOTES
Well, you will appreciate that Southern Sudan is not yet a major
tourist destination, but it is a major destination for aid workers:
that’s the market. To entertain them, an international shopping mall
has been built alongside. The aid agencies themselves have mean-
while spent their time squabbling over which agency should control
the money: every agency wants to coordinate, and none wants to be
coordinated. Currently the government of Southern Sudan is not
sovereign: it shares sovereignty with the federal government of Su-
dan. But in 2011 there will be a referendum on full independence.
Prepare to welcome the new country of Southern Sudan onto the
world stage.
Now take Burundi, another long civil war recently settled.
According to the terms of the settlement imposed by the interna-
tional community, the peace was rapidly followed by an election.
The most extreme among the various Hutu rebel movements won.
Its early acts included imprisoning and torturing its opponents, em-
bezzlement of the public purse for the purposes of importing guns
for a private militia, and expulsion of United Nations peacekeeping
troops. There was nothing the United Nations could do except to
organize its withdrawal.
Now take Eritrea. Eritrea started its post-conflict independence
from Ethiopia with the sort of rave international ratings of which
other African governments could only dream. According to one in-
vestment rating Eritrea was going to be Africa’s Singapore. Within
&
nbsp; a decade it was back at war with Ethiopia, followed by a coup by
the president against his own government, half of the ministers of
which were jailed. Military spending remains on a war footing with
mass conscription. As I write, Eritrea has just expelled the peace-
keepers guarding the buffer zone, not an encouraging step.
And finally the post-conflict darling: East Timor. This is the he-
roic little place that gained self-determination from Indonesia after
a thirty-six-year struggle during which, due to the folly of President
Sukarno’s successor, President Suharto, it was turned into a colony
Inside the Cauldron
79
instead of welcomed as part of the nation. It joined the ranks of the
international community to a chorus of congratulation. It is perhaps
ungracious to point out that if every group of eight hundred thou-
sand people was granted the right to self-determination, the world
would have around eight thousand countries. In other words, inde-
pendence for East Timor did not pass the test proposed by the moral
philosopher Immanuel Kant: “What if everybody did that?”
But, never mind; how has heroic little East Timor progressed
since its independence in 2001? East Timor was one of the 40 per-
cent of post-conflict situations that did not make it through the first
decade without a reversion to violence. In 2006 one of its leading
politicians was found to be importing arms for his private militia.
A large disaffected group in the army that came from the western
part of East Timor attempted a coup and then retreated into the
mountains: sure enough, the same mountains where the civil war
had been fought. In the ensuing struggle a tenth of the population
was displaced. Had not two thousand Australian troops promptly
arrived to put down the coup, a prolonged civil war might have
led to the entry onto the world stage of a new sovereign country of
West-East-Timor.
D e s p i t e i t s i m p o rta n c e , u n t i l r e c e n t ly I had shied off trying to investigate what determines whether a post-conflict peace
endures. In statistical terms it is a difficult question because of the
relatively small number of pertinent observations. By 2006 we had
accumulated data on sixty-six countries, which was at last sufficient
to be worth investigating. This time my team was Anke Hoeffler
and Mans Söderbom, a very smart Swede. We decided to cast our
net wide and investigate on an equal footing all the possible influ-
ences on the duration of peace: political, social, economic, and mili-
tary.
Let’s start with where we left off: democracy and elections. The
80
WARS, GUNS, AND VOTES
standard approach of the international community to the end of a
civil war is to insist on a democratic constitution and crown this
after a few years by an election. This is the theory of legitimacy and
accountability at its clearest. Peace will be secured by an election be-
cause the winner will be recognized as legitimate by the population,
making violent opposition more difficult. Not only will the elected
government be recognized as legitimate, the democratic process will
ensure that it will need to be inclusive and so there will be less reason
for grievance: the government will be accountable to its citizens. It is
time to look at the evidence.
We first checked whether the type of polity affected whether
a post-conflict country reverted to violence. Again we used the
twenty-one-point scale of the Polity IV index, searching along it to
see whether any part of the range was significantly safer than any
other. We did not like what we found. There was a portion of the
range that was significantly safer, but it was the range of intense
autocracy: between –10 and –5. For the countries in this range the
risk of reversion to conflict was much lower: not 40 percent, which
was the overall average, but around 25 percent. Correspondingly,
the polities that were less repressive, that is, with a score of –4 or bet-
ter, had an above average risk of reversion to conflict: not 40 percent
but an astounding 70 percent.
To think concretely, and to take examples that occurred suf-
ficiently recently not to be driving the results, in the early years of
the new millennium both Angola and Sri Lanka made it to peace.
Angola continued to be one of the most repressive regimes on earth,
whereas Sri Lanka was a long-established democracy. The peace
in Angola has held firm, and I expect that it will continue to do
so. The peace in Sri Lanka has already fallen apart: rich-country
governments have heaped the lion’s share of the blame on the Sri
Lankan government rather than on the Tigers, just as they tended
to blame the government of Colombia for the resumption of the war
against the FARC, and the government of Uganda for the running
Inside the Cauldron
81
war against the Lord’s Resistance Army. I am ready to admit that all
three of these governments have probably made mistakes, but what
is manifest is that all three of them are saintly when compared with
that of Angola. In other words, a more democratic polity does not
necessarily make peace more likely.
So much for the effect of the polity. We pressed on to the effect
of elections, introducing them into our model of risks during the
post-conflict decade. There were plenty of elections, but at first we
could not make sense of them: there seemed to be no clear effect at
all. Surely, in the highly charged environment that is typical of post-
conflict, the key political event of an election could not wash over
the society leaving no significant effect. And then we hit on it. A
post-conflict election shifts the risk of conflict reversion. In the year before the election the risk of going back to violence is very sharply
reduced: the society looks to have reached safety. But in the year
after the election the risk explodes upward. The net effect of the
election is to make the society more dangerous.
Why do post-conflict elections have this effect? Well, at this
point we have to leap off the statistical results and start to speculate.
Here is my guess. In the run-up to an election there is a strong incen-
tive for the parties to participate: after all, this is the route to power.
So energies get diverted into campaigning and so risks fall. But then
comes the election result. Someone has won, and someone has lost.
Of course, if this was a genuine democracy the winning party would
say the sort of things that winning parties usually say in genuine de-
mocracies: we will govern on behalf of all the people. And because
of checks and balances that constrained its power while in office, it
would more or less have to do so. If it was a genuine democracy the
losing party would say the sort of things that losing parties usually
say in genuine democracies: we congratulate the winner and will be
a loyal opposition. Because of the restraints on abuse of power, the
losing party would know that it still had a good chance of at
tain-
ing power within five years. Post-conflict situations are not usually
82
WARS, GUNS, AND VOTES
like this. The winner gleefully anticipates untrammeled power: no
checks and balances here. The loser anticipates its fate under the
thumb of its opponents and knows there is but one recourse: back
to violence.
Let’s go back to the first post-conflict situation I described: the
Democratic Republic of the Congo. President Mobutu had been
ousted by the rebel leader Laurent-Désiré Kabila, with the military
backing of Rwanda and Uganda. In 2001 Kabila had been assas-
sinated and his young son Joseph inherited the throne, indeed be-
coming the youngest head of state in the world. I apologize for that
slip: “throne” might give the entirely wrong impression that the true
purpose of the ostensibly Maoist rebellion had been to install an ab-
solute monarchy. Let me at once correct that impression: the young
Joseph was appointed as the next president by due constitutional
process. Indeed, since the international community held the key
cards in this situation, they called the shots, other than the one that
got Laurent-Désiré. Recall that the government was up to its eyes in
debt, was chronically short of revenue, and lacked an effective army.
So President Kabila II had little choice but to acquiesce in holding a
post-conflict election.
The election was to be in two rounds, somewhat like that in
France, with the second and decisive round set for October 29, 2006.
The international community was sufficiently confident of the le-
gitimacy and accountability model that it set the date for the with-
drawal of its peacekeeping forces as October 30, 2006. This was the
denial of reality at its most absurd: democrazy in action. If our re-
sults were right, in one sense the strategy of the international com-
munity was understandable. If events in the Democratic Republic
of the Congo ran to form, the year before the election would be re-
markably peaceful, creating the impression that the society was now
over the period of high risk. Since international peacekeeping is both
enormously expensive and highly unpopular with electorates in the
high-income countries that provide the troops, there is strong pres-
Inside the Cauldron
83
sure to “bring the boys home” as soon as there looks to be no further
Paul Collier - Wars, Guns, and Votes Page 10