The choice will be between peaceful radical decentralization—possibly including secession—on the one hand. On the other hand will be a ratcheting up of coercion and enforced "unity" to ensure political centralization. 

The speed with which American polities proceed toward this situation will depend on two factors, both of which are already at play and contribute to ongoing talk of "national divorce" in the United States. The two factors are:

1. A growing ideological and geographic bifurcation between populations.
2. Growing centralization of state power which raises the stakes for competing cultural and ideological groups. 

The first factor has been illustrated by at least two different developments. The first is that described in Bill Bishop's book, The Big Sort, in which Americans are increasingly moving to states and cities that reflect their own ideological and cultural views. This has apparently even accelerated in the wake of the Covid Panic.   But this is also found in non-geographic factors, and recent polls, including this one from Gallup, which prompted Gallup's researchers to conclude that Americans are increasingly polarized: 

As partisans have become increasingly polarized ideologically, so too have the candidates elected to public office representing those parties. That leaves less room for across-the-aisle negotiation on key issues between the two parties in federal and state government. It also has led to intra-party disagreement between ideologically extreme and centrist officeholders of the same party, which has sometimes made it challenging for the controlling parties of institutions to pass legislation their parties favor or handle basic government functions.

This brings us to the second factor. While ideological division has increased, the size and scope of the central government has grown. This provides the central state with vast powers to engage in immense transfers of wealth that are likely to increase ideological discord. For example, defense spending—a transfer of wealth from current taxpayers to select war-related interests —is now at higher levels than old Cold War peaks. Overall, the central government now spends taxpayer money at unprecedented levels, racking up multi-trillion dollar deficits, even in peacetime. Wealth and income transfers can take on sizable cultural and demographic elements as well. For example, the central government now can subsidize immigration at unprecedented levels—thus increasing internal demographic conflict. 

This spending of all types leads to more price inflation—which largely redistributes wealth upward—and more redistribution of taxpayer wealth and income in general. In a modern democracy, the scope of this redistribution reflects the extent of the important relationship between powerful political interest groups and their patrons within the central government. This vast power to redistribute wealth makes elections a high stakes event. While elections are exceptionally unlikely to lead to any meaningful declines in the size and scope of the state itself, elections nonetheless affect which interest groups benefit most from electoral outcomes. In other words, when the central state is larger and more centralized, the losing political coalitions—composed of specific demographic, ideological, and economic interests—have more to lose when failing to maintain an electoral majority. As competing groups becomes more polarized, the democratic losers will perceive their loss as more substantial that would be the case under a less polarized population.