Now that exams are over I am restarting this series. Once again, each week will focus on a new dynamic with a game theory model. However, I am reducing focus on civilian killings as the conflict moves to rural and jungle regions. I am also adding a subtitle "how to get people to do what you want". This emphasizes that the revolution only makes more obvious dynamics that recur in all politics: n-player game coalition building, intra-coalition bargaining, strategic cost distribution and others.

Burmese Days: How to Get People to Do What You Want

You are a contender for control of a nation state. You have some set of supporters, who expect both policy preferences and personal rewards should you take power. You face another contender, who also has supporters who also expect personal rewards and policy influence from victory. You have promised the world, but you lack the credibility to pry the incumbents supporters away with promises alone. What do you do?

You apply targeted costs to your enemy's supporters, and distributed costs to your own supporters.

Individuals vary in their personal commitment to political movements, and very few supporters are willing to give everything. Meanwhile the costs of support vary greatly. Even the indifferent will brush aside small costs such as price increases, taxation and low-probability risk of violence. But even strong supporters will retreat when threatened with the loss of their livelihoods, torture or the execution of one's family.

Contenders have limited resources to hurt enemy supporters. The least efficient application is to spread those costs equally over the opposing coalition (OC). Individual OC members will barely notice, let alone change behavior. The most efficient is to apply heavy costs to a fraction of the OC, such that they change behavior to supporting you or withdrawing. Then gather more resources and strike at a new faction. This effect was first recognized in American environmental policies, but it applies to all political conflicts.

An American example: After the Jan. 2021 capital riots many ardent Trump supporters faced a very concentrated cost: imprisonment. Strong activist organizations disperse these costs by paying for legal fees and compensating families. The Trumpist faction lacked the institutional support to assist activists after Trump lost power. During his sentencing, one proud boy stated](

I've followed this guy for 4 years and given everything and lost it all (...) my good friends and myself [are[ facing jail time because we followed this guys lead (...) F— you trump you left us on [t]he battle field bloody and alone

That person will not be attending Trump rallies in 2024. Always distribute costs among your supporters. Always concentrate costs on your opponents supporters.

This post will briefly review events in Myanmar, then follow the war of targeted costs through several revolutionary episodes.

Developments since March, in brief

  • The Junta gradually escalated the number of civilian killings, demobilizing mass protests and terrifying opponents off the streets. Some 800 have died.
  • The Junta attempted to regain local government via appointed loyal district chiefs. These chiefs would provide the information to target opposition organizers and project stability/strength. Several appointed chiefs were killed in retaliation.
  • Thousands of Burmese have fled the major cities, as occurred in past crackowns.
  • The Kyat (currency) has lost 19% of its value since the coup. The banking sector is in crisis and withdrawals are now severely restricted
  • Fighting with the minority armies is escalating, but the Tatmadaw (military junta) has bought off critical rebel militia.
  • Two Chinese FN-6 Manpads were leaked into Karen rebel hands, destroying one Tatmadaw attack chopper. It is suspected that China or their proxies leaked the missile system, although the Karen Liberation Army has said nothing.

Targeted costs for thee, distributed costs for me

This section gives three theaters of the targeted costs conflict, in which contenders focus on isolated members of opposing coalitions to force defection.

The banking conflict


The coup created three problems for Myanmar's banks. Bank employees joined the civil disobedience movement and ceased working. The value of the Kyat plummeted. And there was a run on the banks, with thousands looking to withdraw accounts to invest in gold.

The Junta immediately sacked the heads of Myanmar's central bank to gain leverage over the private banks. Then they threatened to cancel the accounts of banks that did not reopen their branches by April 29th. Myanmar's largest bank was threatened with nationalization. The threat of nationalization was credible because governments love stealing things, compromising the largest bank (speculation). With one bank reopening, the smaller banks were isolated and easier to target.

Regrettably, the banks complied and issued their own ultimatum to striking workers; return by the 29th or be sacked. About 80% of workers returned.

Unfortunately confidence in banking has not returned. Customers are withdrawing from the reopened banks but not depositing or servicing their loans. As a result, banking has exacerbated the problems for Myanmar's foreign currency reserves.

The protest conflict

Unlike past urban social movements, the Tatmadaw did not begin their response with mass casualty events. They were likely restrained by either fear of mutiny or pressure from the Chinese.

Hurting even killing protestors is surprisingly inneffective if activists can redistribute the costs among themselves. Activists organize compensation for martyrs in the form of social status through extravagant funerals, financial compensation or simply distribute the costs by moving in crowds. Killing the local butcher in a crowd of ten thousand angers a community but does not disrupt the cost distribution process.

In March it was obvious the Tatmadaw could not find and attack protest organizers, as they searched civilian apartment buildings and indiscriminately arrested protestors. Instead of mass casualty events, the Tatmadaw waited and gradually escalated violence against demonstrators such as tear gas and beatings. As the crowds released the remaining demonstrators were younger, more radical and more willing to resist violently in Palestinian-style nonlethal tactics. The regime then escalated lethal violence against these smaller crowds in single-day killing events starting in tens and reaching a one-hundred-killing day.

This tactic works because only the most organized and motivated activists remain protesting after months of low-level violence. But these are the activists that Tatmadaw wants to hurt most. The Tatmadaw could then use their "budget" of killings to target the activists. The flight to the countryside shows their success.

You can watch this process in human detail in a 15-minute BBC documentary.

The military's targeted costs

An astute read will observe that the military is applying targeted costs to many of its own agents. Soldiers risk their lives leaping from helicopters into Chin or Karen villages. Soldiers are asked to kill peaceful protestors, which is deeply uncomfortable. Appointed district chiefs move to opposition dominated suburbs, where they can be killed at any moment. Junta members are asked to work long hours for a regime they no longer believe in and compromise with eachother. At each step the agent may decide the costs are too high.

The most important tool for solving this is money. We know that the promise of post-crisis reward motivates repressive activity because we can observe when the promise ceases to be credible. In 1978 the Shah of Iran was suffering from terminal leukemia, and dead men do not keep promises. Repressive effort dropped off and the opposition came to power. This suggests that the promise of financial rewards are motivating much of the Tatmadaw's repressive effort.

If the revolutionaries can convince repressors that such financial rewards are not forthcoming, the Tatmadaw would have to demand target costs of its agents, with no distribution. This will trigger a defection cascade, unless pure fear holds them together.

Conflict therefore focuses on sources of foreign currency. Deurbanization has killed the manufacturing export sector, leaving the mineral sectors of jade, gold, natural gas and oil transport forex lifelines to the Junta. These minerals are already targets as Karen rebels have attacked jade mines and debtors have ceased repayments. More on this to come.

Appendix: Players

The Tatmadaw: Myanmar's army, who have held power alone or in coalition since the 1960's. They have a reputation for ruthlessly degrading the citizen organizing capacity. Unlike most Asian despots they are widely hated by elites such as the Buddhist monks, business leaders, the bureaucracy, urban citizens, and ethnic minority armed groups. But they have the minerals and the guns. They currently control the commanding heights of the state.

The NLD: The National League for Democracy is the strongest opposition party, commanding ~75% of the vote in recent parliamentary elections. Their base is in the Bama majority ethnicity, in the business and urban classes. Their leaders are currently in prison, exile or have joined minority rebel armies.

The KLA: The Karen Liberation Army is a rebel army currently fighting the Tatmadaw

New Comment
1 comment, sorted by Click to highlight new comments since:

Most Fantasy RPGs teach similar lessons: concentrate your dps on opponents, and distribute their DPS among your tanks.  This is due to threshold effects - smaller amounts of damage may annoy your opponents, but it doesn't make them ineffective.  Enough damage takes the opponent out of the fight.  

For most of politics and human interaction, this is NOT good advice - there are often compromises or network effects that make all contestants better off if they don't go all-out in hurting anyone.  Once something has reached physical violence levels, though, most of those constraints are ineffective.