Strategic time and decision rents

02 Nov 2026

As a grad student, I was very interested in dynamical systems and game theory.¹ So, I found “dynamic games” mildly annoying as a term: here were games that claimed to be dynamic but didn’t actually have any notion of time!

Except that’s not really true. Dynamic games like the Stackelberg model do have a notion of time. It’s just not the same kind of time that shows up in differential games.² I think of the kind of time that’s relevant to the latter as “calendar time”, and the former as “strategic time”.

Unfortunately, I don’t actually have a concise definition of strategic time. In principle it’s about the sequence of play, like who moves first in a game (hence “first-mover advantage”). But that’s always felt a bit squishy and unsatisfying. Someone almost always decides first. So if we defined strategic time such that the person who decided first is always a leader in some leader-follower game, it would seem simultaneous-move games like the Cournot model are kinda pointless. But we use simultaneous-move games a lot, maybe even more frequently than we use sequential-move games.³ This suggests to me that being a Stackelberg leader isn’t trivially reducible to “the player that decides first in calendar time”.

Since sequential-move games feature strategic time, it’s helpful to think about what the equilibrium conditions of such a game actually say. In a Stackelberg game, player $F$ maximizes their objective taking player $L$’s solution as given, and player $L$ maximizes while anticipating and internalizing $F$’s response. We call $L$ the leader and $F$ the follower. Being first in calendar time can help $L$ establish leadership, but $L$ could just as well make a credible commitment to behaving a certain way and execute second in calendar time. So, strategic time isn’t about time at all; it’s about whose decisions constrain whose. If we dropped the dynamic/temporal language, we might call it something like strategic hierarchy.⁴

Forced to decide

While it’s convenient to model the order of play as though it’s some fixed thing handed down from on high, in reality people spend a lot of effort trying to be a leader and shape play to their advantage.

In business, these efforts can sometimes involve innovations that put a firm on a shorter strategic timescale than its rivals. A shorter strategic timescale doesn’t necessarily mean making decisions more frequently or faster; it may even mean making fewer decisions or getting to costlessly take your sweet calendar time with them. For example, a firm may have a technology that lets it produce at lower cost and greater scale than its rivals, forcing them to plan around its output. Or a firm may try to shorten its strategic timescale by making credible commitments that either reduce the number of remaining decisions it needs to make or increase the number of decisions its competitors must make.⁵ Basically, the shorter an actor’s strategic timescale, the higher it is in the strategic hierarchy and the more competitors have to take its choices as given.

I’m using “strategic timescale” a bit loosely here—what’s the measure?—but I think one could formalize a notion of duration in strategic-time units by considering the number and complexity of decisions a player has compared to their competitors and the potential upside those decisions create. More complexity isn’t always bad; a leader may have a more complex optimization problem in a formal sense, but the resulting payoff gain can be well worth it. Similarly, a follower that must repeatedly make simple decisions with costly measurement steps and constrained or worsening alternatives is on a longer strategic timescale.⁶

Government as a strategic accelerator

Anyway, efforts to reduce one’s strategic timescale may also involve government support. A firm that receives differential funding to develop a technology with a guaranteed schedule of government purchases may end up with lower costs and greater certainty over its revenues, enabling it to out-raise its rivals in private capital markets. Support isn’t just money though; it can include in-kind contributions like expertise from government scientists or engineers, broader policy changes to enable their activities, or even promoting narratives that make it easier for that company to lead the market.

The key thing is that the support is asymmetric: a specific firm gets it and its competitors don’t. I’ve come to think of the value the firm derives from improvements in its strategic timescale as decision rent. In the example above, the additional private funding the favored firm gains (compared to what it could have raised without that support) can generate decision rents if it enables the firm to make more credible commitments or otherwise change the structure of the game in its favor.

Decision rents aren’t limited to Stackelberg games, but I think their nature requires (creates?) a leader-follower structure. There has to be some discrete asymmetry in how players treat each other’s decisions, which I think implies a strategic-time dimension to the game.⁷ A continuous-time Cournot game doesn’t necessarily do this even if there are other asymmetries, e.g., one firm has lower costs than the other.⁸

Don’t challenge me or my national champion again

To be clear, there can be perfectly valid reasons for governments to give asymmetric support and create decision rents. Sometimes there aren’t that many options on the table and a government wants some capability developed For Reasons, so they pick the horse they have. If firms enter to provide the capability later, the supported firm may have decision rents. The government may have tried to select a few firms, and only one succeeded; perhaps there are economies of scale that naturally limit competition, or ways that government support lends credibility to commitments. Or perhaps there is some dimension of geopolitical competition where a dominant champion is preferred to a more raucous and varied ecosystem that does not dominate the global market as strongly.

An economist of Austrian leanings might respond to all this with something like, “yes, government is everywhere and always the source of market power and monopoly rents”. But I think this is something a bit different. Yes, decision rents may beget market power or monopoly rents. But even if the equilibrium of the game doesn’t result in a monopoly, the leader can gain payoff by forcing followers to incur greater decision-making costs. Continuing the private capital example, a firm that secures purchase commitments from a budget-constrained government agency may gain decision rents if its competitors are forced to find other sources of demand to raise similar levels of private capital. If that demand is dispersed across multiple agencies in multiple countries, the followers may have a substantially harder economic problem than the leader; they have to assemble a more complicated set of commitments around what the leader has already secured. Government actions may thus create decision rents even when the government isn’t intending to create a monopoly, or even when it’s trying to create competition.

As interesting as a good perversity thesis can be, I think it’s more interesting when welfare-concerned governments intentionally create decision rents. This seems to be more common in international arenas than domestic ones; it’s easier to proudly champion “our side”. While it’s not conceptually hard to estimate the magnitude of decision rents once you’ve decided they’re there—just compare the leader’s profits when they’re a leader vs a symmetric competitor in a Cournot game—it seems harder to establish whether a real-world game admits them in the first place. But if you’re trying to do industrial policy and create some type of national champion, whether a single firm or a cluster of them, shortening your champion’s strategic timescale and creating decision rents may help you get there at a lower fiscal burden than pure subsidy.