07/22/2025
Game of Power: From Dilemma to Deadlock and Its Policy and Investment Implications
By Robert Wright
Application of game theory to American politics, past and present, shows that the “game of power” in Washington recently changed. Long a “Prisoners’ Dilemma” that rewarded long-term cooperation between the two political parties, it has become a “Deadlock” that encourages hyper-partisanship. Policy stability gives way to policy switching and economy-damaging uncertainty as a result.
To refer to political strategy as a game, as in Game of Thrones or House of Cards, is not to trivialize it but rather to think about it in a disciplined way. The game of power is a two-player simultaneous-move non-cooperative game in which the two players, the two major political parties, make strategic decisions knowing the payoff structure but without knowing the other’s move.
For most of U.S. history, the game of power had a payoff structure like in a Prisoners’ Dilemma, which gets its name from the story used to illustrate it. Two detainees under police interrogation face different outcomes (payoffs) depending on their joint actions. The payoff structure can be stated formally as T (Temptation) > R (Reward) > P (Punishment) > S (Sucker) and subject to the further stipulation that 2R > T + S.
To refer to political strategy as a game, as in Game of Thrones or House of Cards, is not to trivialize it but rather to think about it in a disciplined way. The game of power is a two-player simultaneous-move non-cooperative game in which the two players, the two major political parties, make strategic decisions knowing the payoff structure but without knowing the other’s move.
For most of U.S. history, the game of power had a payoff structure like in a Prisoners’ Dilemma, which gets its name from the story used to illustrate it. Two detainees under police interrogation face different outcomes (payoffs) depending on their joint actions. The payoff structure can be stated formally as T (Temptation) > R (Reward) > P (Punishment) > S (Sucker) and subject to the further stipulation that 2R > T + S.
Prisoner’s Dilemma Payoff Matrix
|
Prisoner A/Prisoner B |
Cooperate |
Defect |
|
Cooperate |
R, R |
S, T |
|
Defect |
T, S |
P, P |
The payoff matrix indicates that if both prisoners cooperate with each other (neither rats the other out), both gain amount R, a reduced sentence. If one defects (rats) and the other cooperates (keeps quiet), the latter will be severely punished (S) and the rat will get off with a wrist slap (T). If both prisoners rat (defect), both get stiff sentences (P) but not as harsh as the one that will be imposed if the other player rats.
If played once, and not embedded in a larger game like “snitches get stitches,” the rational strategy is to defect. If the other player cooperates, the rat gets off easy. If the other player rats, much is lost by cooperating. So, both sides expect the other to rat and do so themselves.
That might seem like an arcane thought exercise, but the Prisoners’ Dilemma tradeoff structure occurs in many real-world situations, including economic exchange. Thankfully, trading between the same two parties often recurs indefinitely and that breeds cooperation. As both players seek to earn R (which recall is larger than P), the mutual gains from trade proceed at every turn.
For a long time, America’s political parties faced a Prisoners’ Dilemma payoff structure in a recurring game. Using numbers to illustrate:
If played once, and not embedded in a larger game like “snitches get stitches,” the rational strategy is to defect. If the other player cooperates, the rat gets off easy. If the other player rats, much is lost by cooperating. So, both sides expect the other to rat and do so themselves.
That might seem like an arcane thought exercise, but the Prisoners’ Dilemma tradeoff structure occurs in many real-world situations, including economic exchange. Thankfully, trading between the same two parties often recurs indefinitely and that breeds cooperation. As both players seek to earn R (which recall is larger than P), the mutual gains from trade proceed at every turn.
For a long time, America’s political parties faced a Prisoners’ Dilemma payoff structure in a recurring game. Using numbers to illustrate:
Prisoner’s Dilemma in Washington
|
Republican/Democrat |
Cooperate |
Defect |
|
Cooperate |
40, 40 |
20, 50 |
|
Defect |
50, 20 |
30, 30 |
As in the previous prisoner and trade examples, with that payoff structure long-term cooperation is the best strategy for both parties, who can expect to earn 40 (billion dollars, million votes, bills passed, whatever) each time they play. If the two parties colluded such that they took turns defecting while the other cooperated, each party alternates earning 50 and 20 each time round, for an average payout of 35 each. If they always defected, they would earn just 30 per round. Earning 40 is obviously better than 30 and also better than the 35 resulting from alternating between 50 and 20.
From the nation’s founding until recently, both major political parties routinely cooperated with each other on important policy issues. Bipartisanship and general policy consensus was so common that some Americans complained that a “uniparty” or a “deep state,” not competitive parties, controlled government. Political satirists joked that elections pitted nearly identical candidates against each other, like “Frick” against “Frack,” “Kang” against “Kodos,” or “Jack Johnson” against “John Jackson.”
Cooperation, however, brought with it policy stability and hence a relatively high degree of investment certainty. Democrats created the Social Security system, for example, but Republicans under President Eisenhower extended rather than dismantled it. Democrat Jimmy Carter started the deregulation wave that presidents from both parties continued until century’s end. The Overton Window, the range of acceptable policy reforms, was open, but just a crack.
Today, by contrast, the Overton Window is wide open. Policy discussions swing from modern monetary theory to a gold or even bitcoin standard. Forcing out the head of the Fed, once unthinkable, is now openly discussed. Presidents now regularly cancel each other’s executive orders with their own and openly disavow Supreme Court decisions.
As the following chart shows, since 1985 policy uncertainty increased somewhat when control of the government shifted from one party to the other. During the most recent changeover, however, it spiked to a level seen only once before, during the Covid pandemic.
From the nation’s founding until recently, both major political parties routinely cooperated with each other on important policy issues. Bipartisanship and general policy consensus was so common that some Americans complained that a “uniparty” or a “deep state,” not competitive parties, controlled government. Political satirists joked that elections pitted nearly identical candidates against each other, like “Frick” against “Frack,” “Kang” against “Kodos,” or “Jack Johnson” against “John Jackson.”
Cooperation, however, brought with it policy stability and hence a relatively high degree of investment certainty. Democrats created the Social Security system, for example, but Republicans under President Eisenhower extended rather than dismantled it. Democrat Jimmy Carter started the deregulation wave that presidents from both parties continued until century’s end. The Overton Window, the range of acceptable policy reforms, was open, but just a crack.
Today, by contrast, the Overton Window is wide open. Policy discussions swing from modern monetary theory to a gold or even bitcoin standard. Forcing out the head of the Fed, once unthinkable, is now openly discussed. Presidents now regularly cancel each other’s executive orders with their own and openly disavow Supreme Court decisions.
As the following chart shows, since 1985 policy uncertainty increased somewhat when control of the government shifted from one party to the other. During the most recent changeover, however, it spiked to a level seen only once before, during the Covid pandemic.
The sea change in party strategy reflects a new payoff structure. Due to increased political polarization of voters and donors, and the erosion of traditional norms and constitutional checks and balances, the parties now face a payoff structure called Deadlock that encourages constant defection.
Again illustratively, mutual cooperation now garners each party 30 instead of 40, while mutual defection earns them both 35. They now earn 55 instead of 50 for defecting if the other side cooperates but get only 10, instead of 20, if they cooperate while the other party defects. The payoff of coordinated swapping now alternates between 55 and 10, for an average payoff of 32.5. Both parties will therefore defect every turn because doing so pays more than mutual cooperation or alternating big wins.
Again illustratively, mutual cooperation now garners each party 30 instead of 40, while mutual defection earns them both 35. They now earn 55 instead of 50 for defecting if the other side cooperates but get only 10, instead of 20, if they cooperate while the other party defects. The payoff of coordinated swapping now alternates between 55 and 10, for an average payoff of 32.5. Both parties will therefore defect every turn because doing so pays more than mutual cooperation or alternating big wins.
|
Republican/Democrat |
Cooperate |
Defect |
|
Cooperate |
30, 30 |
10, 55 |
|
Defect |
55, 10 |
35, 35 |
In sum, the parties are now “deadlocked” in the sense that mutual compromise runs against their interests. The out-of-power party therefore does everything it can to stop the policies of the party in power. If unsuccessful, it will reverse the policies upon its return to power. Policies thus increasingly hinge on election results, which have become increasingly high stakes and difficult to predict. As a result, investors and businesses must consider the risk of rapid policy switches to an extent unseen since the New Deal.
