Chapters 2–4 described the major harms for which, it was argued, state government is largely responsible. But does state government supply important offsetting benefits? In this chapter, I have tried to locate or imagine every conceivable benefit that one might claim state government offers. I didn’t have to look hard. Most of the claimed benefits I could think of, and some that I would not have thought of on my own, have already been asserted by others.

To be sure, these previously invoked benefits have not been packaged as arguments for the retention of state government; serious debate over whether to abolish state government simply has not taken place. Rather, these arguments have typically been presented in defense of federalism. And that is a different debate.Footnote ¹ It is possible to have states or their equivalent without federalism, as the below discussion on decentralization illustrates. Indeed, unitary systems in which the various subunits are subordinate to the national government are plentiful in today’s world. But it is not possible to have federalism without states or their equivalent. Thus, every one of the arguments for federalism presupposes, and therefore is necessarily also an argument for, state government.

As I will show, these arguments are far less convincing than they initially appear. Each of the claimed benefits either does not exist, exists but is very minor, or could be replicated at least as well and often better by some other level of government.Footnote ² A common flaw in most of those defenses of states is that the state is simply too blunt an instrument to achieve the claimed benefits. The local governments, to which I favor transferring many of the states’ functions, will be shown to be far more precise.

Some scholars – most notably Edward Rubin & Malcolm Feeley – have made the different point that you don’t need a federation to realize the benefits often cited for it; decentralization alone, they say, would be just as effective.Footnote ³ I agree but go a step further. You don’t need state government at all to achieve the advantages that either federation or mere decentralization is said to provide. As the discussion in this chapter will show, the same arguments that make mere decentralization preferable to federation make the abolition of state government better yet – as do additional arguments.

Of course, there are many insightful commentators – perhaps most – who do not share my dim view of states’ value.Footnote ⁴ An otherwise thoughtful and highly informative book by political scientists Ann O’M. Bowman, Richard Kearney, and Carmine Scavo overflows with cheery, uplifting tributes to the behaviors and accomplishments of states. On one page alone appear extravagant references to the states’ capacities for “innovation,” their transformations into “viable and progressive political units,” the “positive outcomes” states generate, and the “new directions” they have “creatively crafted.”Footnote ⁵ Negative behaviors are unavoidably acknowledged on occasion but receive scant emphasis. Neither the modern tidal wave of voter suppression and other counter-majoritarian state actions nor the states’ disingenuous defenses of those measures come across.

I concede the obvious: Among the fifty states, all of which have been with us for anywhere between several decades and more than two centuries, and each of which has legislated on thousands of subjects along the way, it is easy to find countless examples of exemplary actions. It is for the reader to judge whether the benefits that state government sometimes brings offset the damage, described in Chapters 2–4, that they have done to our democratic ideals and to the fiscally efficient operation of our multilevel government structure.

A Diffusing Government Power

The most frequently cited benefit of states is that they prevent what would otherwise be an unhealthy concentration of power in the national government. Diffusing government power, the argument runs, enables the federal government and the states to check each other’s excesses, thereby protecting individual liberties. In the words of the Supreme Court, “State sovereignty is not just an end in itself: Rather, federalism secures to citizens the liberties that derive from the diffusion of sovereign power.”Footnote ⁶ The Antifederalists strongly agreed and cited additional benefits of diffusing government power, including their belief that it helps check corruption.Footnote ⁷

In Printz v. United States,Footnote ⁸ the Supreme Court similarly invoked diffusion of government power as a central purpose of our federal system. Interim provisions of the federal Brady Act required the chief law enforcement officers of each jurisdiction to complete certain paperwork, and perform background checks, before firearms could be sold. In its 5-4 decision, the Court declared those requirements unconstitutional. The Court held (among other things) that requiring state officers to assist in federal law enforcement violated the principle of dual federal and state sovereignty. The principle that the federal government may not force the states to administer a federal program has come to be known as the “anti-commandeering doctrine.”

The Court in Printz quoted James Madison:

While much of the Court’s opinion comes very close to treating state sovereignty as an end in itself, the above passage disavows sole reliance on that notion, making one feeble attempt to add content to the supposed benefits of dual sovereignty. The Court – and Madison – say that our federal system gives the people double protection: Each of the two sovereigns will “control” the other, and each will be subject to additional internal controls (the latter being a reference to separation of powers).Footnote ¹⁰

But fears that a unitary system would unleash an untethered national government are less concerning than might initially appear. As a preliminary matter, the abolition of states would end the distortions produced by the Electoral College, by equal Senate representation for states with wildly different populations, and by the endless parade of gerrymandering, voter suppression, and similar actions of partisan state legislatures. Without those substantial counter-majoritarian advantages, a radical, authoritarian trifecta at the national level (both houses of Congress and the White House) would be much less likely to start with.

Admittedly, though, it would not be impossible. So let’s assume for the sake of discussion that, against the odds, such a government achieves precisely such a national trifecta even without the unfair advantages that flow from the status and the actions of state governments. Even then, there would be several reasons to discount either the need or the effectiveness of federalism as an antidote.

First, the diffusion benefits of federalism sound more consequential than they are. Yes, federalism diffuses power in the sense that the federal government handles some subjects and the states handle others. But within each of those two orbits, one level of government or the other will always have a virtual monopoly of power. When the federal government exercises a power that the Constitution gives it, there is nothing that federalism enables the states to do about it. Conversely, when a state exercises a power that the Constitution doesn’t give to the federal government, there is nothing the federal government can do about it. Within each of the two spheres of operations, therefore, federalism does little or nothing to check the exercise of government power. And even when the federal government sees fit to partner with the states in the implementation of a federal program or otherwise share a particular governance function with the states, the supremacy clause ensures that the federal government will always have the last word. If it wants to be tyrannical, there’s nothing the states can do to stop it (other than go to court, which either private citizens or, in my proposed unitary nation, local governments could also do). So while federalism means that the various subjects of governmental regulation will be divided up between these two levels of government, the fact remains that, with due respect to Madison and the Supreme Court, on any given subject neither level of government will have any meaningful “control” over the other.Footnote ¹¹

I will concede two narrow exceptions to that assertion. Under the anti-commandeering doctrine, even when Congress regulates a subject area that the Constitution otherwise assigns it, states can generally refuse to provide affirmative help in implementing Congress’s wishes. Thus, if the particular program is one that requires ground level implementation, Congress needs to find an alternative to mandating state partnerships. Fortunately, several such alternatives exist. Even in our current federation, one option is voluntary partnerships with willing states, though admittedly that solution offers no assurance of nationwide cooperation and at any rate gives states negotiating leverage. And that leverage can sometimes be constraining. More important, however, with or without states the national government can, as it often does today, administer the law through a network of regional offices of the relevant federal agency. And if state government were abolished, still another option would be to assign the implementation function to local governments, which would remain free to partner with other local governments when it is efficient to do so.

The other narrow exception relates to law enforcement. The recent criminal actions that state prosecutors have brought against former President Donald Trump reveal the potential utility of subnational enforcement authority. If a tyrannical national government were in power (not the case at the time of these state prosecutions), national prosecutors might be less likely to bring criminal charges against the president, other members of the administration, or members of the president’s party in Congress. But one struggles to recount frequent instances in which state police or prosecutors have had to initiate criminal proceedings against federal government officials because the federal government was refusing to do so for political reasons. If anything, the reverse comes to mind more readily; the need for the federal government to send National Guard troops to Alabama to force Governor George Wallace and other state officials to enforce federal civil rights law is not so distant a memory.Footnote ¹²

The good news is that, even if federalism meaningfully constrained central government overreach, we wouldn’t need federalism for that purpose. The central government is already triply constrained even without it: by the constitutional provisions that protect individual rights, by constitutionally enshrined separation of powers principles (a check that, in the light of political parties’ current roles, I do not want to exaggerate),Footnote ¹³ and by the political process (a constraint that, as just noted, should grow stronger once state government and its counter-majoritarian effects no longer distort electoral outcomes). In addition, there is no reason that popular initiatives and referenda could not be introduced at the national level. They would be no harder to administer nationwide than the national presidential vote proposed here and would add another valuable political check.

If even the combination of those multiple constraints is thought to be inadequate, there is yet another whole set of checks and balances – internal separation of powers. This strand of separation of powers operates on a vertical basis within the federal executive branch. Its principal form is review of federal agencies’ significant regulatory proposals by the Office of Information and Regulatory Affairs (OIRA). That office is located within the Office of Management and Budget, which in turn is part of the Executive Office of the President.Footnote ¹⁴

Jennifer Nou points out that OIRA review can be even more important than judicial review. The former is more systematic, less deferential to the agency, and earlier in the process.Footnote ¹⁵ Of course, no one likes to have their decisions reversed, and federal agencies are no exception. As Nou demonstrates, executive branch agencies currently have a variety of ways to insulate their decisions from OIRA review. They can, for example, choose particular regulatory instruments that are not subject to OIRA review at all or find ways to make OIRA’s review more time-consuming or otherwise more costly, in order to discourage rigorous scrutiny. But as Nou proceeds to argue, presidents have ways to resist and to preempt those kinds of self-insulation strategies. That presidential control, while far from absolute, thus remains a valuable safeguard against runaway federal agencies – one that supplements the traditional separation of powers constraints imposed by Congress and the courts.Footnote ¹⁶

Transferring a large chunk of the states’ current responsibilities to the national government would, of course, require a massive increase in the sheer size of the national government. New national agencies would be required. While the management of those individual agencies should be no more difficult than the management of the analogous state agencies they would be replacing, the job of supervising and coordinating the whole would clearly pose additional challenges. This is not a concern about authoritarian or tyrannical government, but I concede that an expansion this significant would make OIRA’s work even more vital. OIRA would need to add new branches, and the jobs of the OIRA Administrator and the president would become less enviable.

OIRA does, however, have help. Congress has found other internal (within the executive branch) ways to cabin federal agencies. When Congress wants agencies to take adequate account of values that are outside of, or even contrary to, an agency’s mission, it often creates what Margo Schlanger has called “offices of goodness.”Footnote ¹⁷ Schlanger limits the term to those offices that are advisory only (not operational), have a mission defined by specific values (e.g., civil liberties or fiscal efficiency), and are internal to and dependent on the agency. Her chief example is the Department of Homeland Security’s Office for Civil Rights and Civil Liberties (which she headed).Footnote ¹⁸ And as Neal Katyal has pointed out, still other devices for checking the actions of executive branch agencies, though not possessing all the attributes of “offices of goodness,” include Offices of the General Counsel, Offices of Inspectors General, Ombudsmen, and the career Civil Service.Footnote ¹⁹

Finally, suppose one disagrees with all of the preceding points. Suppose, in other words, that one does not believe that abolishing all the state-related counter-majoritarian distortions of the political process would appreciably diminish the chances of a radical authoritarian party achieving a national trifecta; that there is some way in which federalism enables the states to control the national government even when the latter is otherwise exercising one of its constitutionally enumerated powers; and that all the existing constraints on the federal government, even taken collectively, are insufficient. If one therefore believes that dividing functions between the national government and fifty states offers the essential benefit of diffusing power, then imagine how much more diffusion we would get by dividing the functions between the national government and tens of thousands of local governments, as this book proposes.Footnote ²⁰ That’s diffusion on steroids.

True, one might reply, these functions could be taken on by the local governments. But if the enumerated powers doctrine is repealed, the national government wouldn’t have to let them do these things. And, as the antifederalists worried, the natural tendency of those in positions of power is to look for ways to expand their power, not reduce it.

Nor, I must acknowledge, is this fear of a runaway national government just a concern of states’ rights conservatives. What if the Electoral College were to install another Donald Trump as president and a similarly inclined or compliant Congress controlled by that president’s party is also elected? If there are no states, and if the national government is no longer limited to enumerated powers, what prevents the latter from usurping subject areas best left to local governments? The constitutional provisions that guarantee certain individual rights and require separation of powers will continue to impose limits, but not on the national government’s choice of subject areas to regulate. It could prohibit or severely restrict the zoning authority of local municipalities, or mandate weekly garbage collections, or decide which street corners should have stop signs. It could make it a criminal offense to teach elementary or secondary school students about, or even to mention, slavery or voter suppression or protests against particular wars or other material deemed “unpatriotic” – or other taboo topics such as evolution or climate change or unionization or sexual orientation or nonbinary gender or contraception or abortion.Footnote ²¹

That would be the worry, and it is a legitimate one. As just noted, a unitary system would reduce the chances of a radical party monopolizing the political branches of the national government in the first place, but such an event would still be possible.

First, however, let’s be clear about the scope of the concern. Eliminating the enumerated powers constraint would have no effect on any national actions that are already within its current enumerated powers. The only effect of lifting that constraint would be on those national actions that the enumerated powers doctrine currently precludes. Those powers are not trivial, but at least the broad powers already exercised by the national government would not be expanded.

Second, while the chances of a future national government impinging on local government in one or more of the ways the argument hypothesizes are not zero, they are not high. In some of the examples, such as local decisions on zoning, garbage collection, or street signs, one can safely assume the national government would have little interest. In most of the other examples, feared federal excesses would be expected to enrage local citizenries. The political costs would run both wide and deep. For those, the political process – especially one shorn of its counter-majoritarian blemishes – would offer the best medicine.

Finally, the current constitutional scheme gives the states just as much leeway to operate in those subject areas that are outside the federally enumerated powers as the federal government has to operate within them. In virtually all the examples given, one can say that if the federal government has no business imposing the particular policy on local governments, the state government doesn’t either. For anyone who views any of those hypothetical directives as undesirable (and I certainly do), it’s a question of “pick your poison.” When it comes to the kinds of extreme measures that this defense of federalism conjures up, my view is that, if anything, it is state governments that pose the greater threat.

That is so because, without discounting the rightful fear of a radically authoritarian national government – we had one as recently as 2017–21, though only with the aid of such state-related counter-majoritarian institutions as the Electoral College and the Senate – radical state governments willing to intrude in the ways feared are far more likely. There are solid red states and solid blue states, but there is no solid national state. Together, the tightness of many nationwide elections and the knife-edge balance of power that so often grips one or both houses of Congress naturally push the federal government into political terrain somewhere between the reddest states and the bluest ones. As long as there are state governments that need answer only to populations in which solid majorities are either reliably progressive or reliably conservative, the political headwinds against radical state intrusions are far weaker than those faced by a national government that answers to a more heterogeneous, and more fluid, countrywide population.

This last point might initially appear to undermine my proposal to shift substantial power from the states to their local municipalities. After all, just as the national government is likely to be less extreme than the most extreme state governments, so too a given state’s government is likely to be less extreme than the most extreme of that state’s municipalities. Under the current system, states can check overreaching local policies; without states, only the national government would be able to do so. And if a radical national government were sympathetic to those local policies, it would be unlikely to step in.

I concede that danger. But the alternative poses a corresponding danger. Under the present system, whole states can – and often do – run roughshod over the decision-making authority of their local governments, as Chapter 3, Section F highlighted. In a unitary nation, that would no longer be the case.

The preceding analysis is addressed to one specific problem – a radical, authoritarian national government ready to impose its views nationwide and against the wishes of the majority. That government might or might not sincerely believe it is acting in the best interests of the nation. In contrast, in Federalist 10, James Madison considered a different but parallel worry – “factions.” By this term, he was referring to groups of people willing to pursue their own self-interest even when doing so harms other citizens or the nation as a whole.

Madison maintained that a large republic, headed by a strong national government, is best situated to control such factions. His argument began with the premise that there’s a certain optimal size of a legislative body – too large and it becomes confusing, too small and it can be disrupted too easily by legislators of bad will. That optimal size – or at least optimal range – doesn’t vary appreciably with the size of the governed population. Therefore, in a large population, there are more good candidates to choose from, for roughly the same number of spots. In addition, “as each representative will be chosen by a greater number of citizens in the large than in the small republic, it will be more difficult for unworthy candidates to practice with success the vicious arts by which elections are too often carried.” Moreover, he added, “Extend the sphere, and you take in a greater variety of parties and interests; you make it less probable that a majority of the whole will have a common motive to invade the rights of other citizens.”

For those same compelling reasons, controlling a politically radical minority – whether or not it is motivated by self-interest – is easier in a large population unit like the nation than in a smaller population unit like an individual US state. Again, we have solid blue states and solid red states, but not – at least yet – a solid national state.

A slight variation of the diffusion of power argument is “What about the ‘good’ states?” Today, if the national government is bad, at least the good states can enact laws that benefit their own citizens.

Good and bad, of course, are in the political eyes of the beholder. My normative view is that, whatever one’s general political leanings, all Americans of good faith should be in favor of reforms that restore democracy and reduce fiscal waste. Those ought to be consensus views. But I recognize that the reality is different. Today, many Americans are willing to sacrifice either of those benefits to achieve competing goals to which they attach higher priorities.

At any rate, this point cuts both ways. Yes, one who dislikes the national government will be happy that states can offset some of its policies, at least for their own citizens. But that coin has a flip side. One who likes the national government typically won’t like the prospect of states undermining its policies.

Finally, as the next section elaborates, even in states that we think of as solid red or solid blue, the popular political preferences almost always vary dramatically from one area of the state to another, especially as between the urban and rural areas. Thus, in a unitary republic, if you don’t like the national government you can still take solace in the fact that there will be many local governments in a position to adopt more agreeable laws. Moreover, unlike the case today, in a unitary system a local government will be so positioned even if it is located in what had previously been a state with polar opposite political leanings and a willingness to impose those ideological preferences on its local governments.

One last point deserves mention. Jessica Bulman-Pozen has argued that partisan motives frequently inspire states to push back against federal action.Footnote ²² As she makes clear, her point is mainly descriptive, not a normative argument for partisan federalism, which could be seen as either beneficial or harmful.Footnote ²³ My view is that, even if these sorts of pushbacks are viewed as a beneficial check on the federal government in the system as it is currently configured, they would not be needed in a unitary American republic. At least for those pushbacks that take place in court, challenges to the federal government’s authority could be brought either by the local governments to which many of the state powers would devolve, or by aggrieved private individuals.

B Distinctive Needs and Preferences

Another benefit often claimed for federalism stems from the state-to-state differences in their populations. Bowman et al. emphasize that “[s]tates and their communities have different fiscal capacities and different voter preferences for public services and taxes.”Footnote ²⁴ The Supreme Court has also invoked that idea in defense of federalism. At least twice, it has said that federalism permits “local policies more sensitive to the diverse needs of a heterogeneous society …”Footnote ²⁵

Those observations have been the basis for two different arguments on behalf of federalism. One is that federalism makes it possible to tailor laws and policies to the specific needs – and, one might add, political preferences – of various, diverse state populations. That democracy-focused argument – honoring the will of the relevant electorate – is often referred to as the “voice” rationale. It is taken up in this section. The other consequence of such tailoring, invoked more explicitly by Bowman et al.,Footnote ²⁶ is that citizens who are dissatisfied with the laws and policies of the states in which they live have the option of moving to a state with policies more to their liking. Thus, they argue, federalism creates a healthy competition among states to retain their citizens. That second argument – citizen choice and the resulting interstate competition for citizens – is often referred to as the “exit” rationale. It will be the subject of Section C.1.

Barry Friedman argues that “[s]tate and local governments can work to protect the safety, health and welfare of their citizens” [emphasis added]. They can often do this more effectively than the national government, he suggests, because they can account for local conditions and needs.Footnote ²⁷ But by lumping state and local governments together, I fear, he gives state government a free pass. I say this for two reasons.

First, and perhaps counterintuitively, it is not a given that in practice state government is any more likely to tailor its laws to the people than the federal government is. Miriam Seifter has persuasively cast empirical doubt on the premise that state government is closer to the people (and thus better able both to tailor its policies to those preferred by the people and to partner with the federal government in the administration of federal laws). In large part, she shows, that is because civil society, while providing a strong check on the federal government, has been a relatively weak check at the state level.Footnote ²⁸

But second, even if, contrary to Seifter’s argument, states are indeed better positioned than the federal government to tailor their laws and policies to the interests and political preferences of the various state populations, their ability to do so is rooted in the fact that states outnumber the national government 50 to 1. Their populations and their territories, therefore, are only fractions of the national population and territory. But local governments in turn outnumber the states 90,000 to 50, so if anything they can achieve the tailoring objective even better than states can.

In fact, Friedman’s main examples of functions for which states and local governments are better suited than the federal government are instructive: “How many police shall there be, how shall they conduct themselves, and where shall they be stationed? Where shall schools be, and what shall they teach? Should we have light rail, or other means of assisting commuters?”Footnote ²⁹ All of these are quintessentially local government decisions, rarely if ever laws that apply uniformly statewide. He then goes on to assume that without federalism these decisions would have to be made by the national government, which he (and I too) believe to be unsuited for those tasks. But it is hard to imagine Congress or federal agency officials having any interest in making those decisions, and he does not consider leaving them with local authorities even though that is where the cited decisions typically rest today.

The larger point is that, in some subject areas, one could reasonably posit a democratic norm that prescribes tailoring laws and policies to the political preferences of the relevant subnational constituencies. The question that norm begs, however, is which subnational constituency is optimal – the statewide population or the various local populations.

As to that, the reality – both historically and today – is that intrastate polarization eclipses interstate polarization. The vivid divide between urban America and rural America has been thoroughly documented by Jonathan Rodden and others.Footnote ³⁰ I would be hard-pressed to identify any cultural, racial, political, or other similarities between urban and rural populations, other than at a level of generality so high as to leave them indistinguishable from residents of any other state. Ask Philadelphians with whom they have more in common when it comes to interests, needs, lifestyle, political views, party, or anything else: residents of Camden, who live right across the state line in New Jersey, or farmers in western Pennsylvania. Conversely, ask farmers in western Pennsylvania with whom they have more in common: other farmers who live across the state line in eastern Ohio, or their fellow Pennsylvanians who live in Philadelphia. The answers are not hard to guess.

Happily, we don’t have to guess. Just look at the 2020 presidential election results in almost any state. Missouri and Illinois provide nice illustrations. In Missouri, Trump beat Biden by 15 percentage points statewide.Footnote ³¹ In Illinois, Biden beat Trump by 17 percentage points statewide.Footnote ³² By any measure, those are substantial margins, and it seems fair to describe these two states today as reliably red and reliably blue, respectively.

And yet, impressive as those victory margins were, they absolutely pale when compared to the differences in voting patterns within each of those states. In Missouri, Biden carried the City of St. Louis by sixty-six points. In the heavily rural counties located in the four corners of the state, Trump beat Biden by anywhere between forty-five points and sixty-seven points.Footnote ³³

Illinois followed the same urban versus rural pattern. In Cook County, home to Chicago, Biden defeated Trump by forty-seven points. The counties that form three of the corners of Illinois are rural counties, where Trump won by margins ranging from seventeen points to forty-eight points.Footnote ³⁴ In the two rural counties that jut out to the west, Trump’s victory margins were forty-seven points and sixty-four points.Footnote ³⁵

One can construct other arguments for the preservation of state government. They are considered in the other sections of this chapter. But as for the specific goal of tailoring laws and policies to the political preferences of the various subnational constituencies, I cannot identify any credible advantage to statewide tailoring over local tailoring. As the election numbers highlighted above amply illustrate, states are absurdly crude proxies for gauging the political preferences, and presumably therefore the needs and interests, of subnational populations. Thus, states are crude instruments for accomplishing the legitimate goal of tailoring laws and policies to those particularized needs and preferences. The mismatch is also unnecessary, because the local governments, to which many of the current state functions would be transferred under my proposal, are far better proxies for the varying preferences of different populations.

A separate argument for valuing the distinctive needs and preferences of individual states’ citizens is that doing so creates a kind of steam valve, relieving pressure that would otherwise trigger internal strife. Antifederalist 7, by “Philanthropos,” went so far as to suggest that for this reason too strong a national government could lead to civil war. One might even assume that it did indeed do so, though in that instance the moral price of avoidance would have been accepting the enslavement of an entire race. At any rate, the civil strife argument too is easily answered, and for the same reason that the democracy-based argument for subnational choice does not justify states: If the states are thought to be useful instruments for placating their citizens by recognizing their distinctive interests, then the local governments, which as noted above are far more precise proxies, would be better instruments still.

While the preceding discussion illustrates why local governments are better suited than state governments to accommodate subnational concerns, there are many areas in which the national government’s ability to respond to diverse needs and interests also exceeds that of the states. As others (noted below) have documented, the states’ differing resources have left them incapable of addressing serious state-to-state social and economic inequalities. In several respects, in fact, state government has only made those problems worse.

One striking example is the subject of a careful empirical study by Jennifer Karas Montes.Footnote ³⁶ For every calendar year from 1970 to 2014, she correlated the average life expectancies of people living in each of the fifty states with the liberality or conservatism of those states in each of eighteen different policy domains. She found that “[m]ore liberal versions of policies related to abortion, civil rights, environment, gun control, immigration, LGBT rights, private sector labor, and tobacco tax, as well as a measure of policy innovation, predict longer lives.” How much longer? “If all states enacted liberal policies across the 18 domains, our study estimated that U.S. life expectancy would increase by 2.8 years for women and 2.1 years for men. However, if all states enacted conservative policies, U.S. life expectancy would decline by 2.0 years for women and 1.9 years for men.” Moreover, widening political polarization in recent years has only increased the gap in life expectancies between states with liberal policies and those with conservative policies.Footnote ³⁷

While the point of that empirically comprehensive article was to highlight the health and life expectancy benefits of liberal policies, it also fortifies the point, more relevant here, made by writers like Donald Kettl and Jamila Michener: State-to-state policy differences generate dramatic geographic inequalities in the health, welfare, and life expectancies of their citizens.Footnote ³⁸ All else equal, your chances of living a long and healthy life depend heavily on the policies of the state you live in. And that is a problem that too many states have proved either unable or unwilling to fix.

Of course, just as economic inequalities exist from state to state, so too they exist from one local area to another. What role the national government should play in addressing geographic economic disparities is an issue that would arise with or without states; it is thus beyond the scope of this book. The only observation I can offer is that, without states, the central government would remain the only actor with the capacity to end those sorts of geographic inequities. Nothing would force Congress to fulfill that responsibility even then, but federalism would no longer provide cover for its inaction.

I can think of one final argument that might be made as to why the surgically targeted policymaking that federalism purportedly permits is beneficial. Members of Congress are supposed to promote the welfare of both the nation and their own constituents. Those two obligations can conflict; what is best for the people of one particular state or one particular US House district is not necessarily best for the country as a whole. State legislators and state officials, patriotic as some might genuinely be, are less burdened by that dual loyalty. They can give top priority to the interests of their respective states.

That distinction, however, might be more theoretical than practical. Unless members of Congress plan to retire, the realities of reelection can never be far from their minds. And that means focusing intensely on the desires of one’s own state or local constituents, just as their state legislative counterparts do.

In addition, the local governments that would assume the lion’s share of the localized functions currently performed by states would have at least as much incentive as state legislators currently have to prioritize local interests. They would likely have even greater incentives to do so, since, as discussed earlier, the political preferences of their local constituents might diverge sharply from those of individuals who live elsewhere in the state.

C Interstate Competition

States, it is sometimes said, are a source of healthy competition. Two different kinds of competition are commonly asserted: Through their differing laws and policies, states give people a choice of where to live. If you are dissatisfied with the laws of your current state of residence, or enamored of the laws of another state, you can vote with your feet. This so-called “exit” option incentivizes states to be responsive to their citizens’ needs and desires. The other strand of the competition argument is that, similarly through their laws and policies, states compete to attract businesses that will benefit their economies.

D Laboratories of Innovation

In 1932, Supreme Court Justice Louis Brandeis wrote: “It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.”Footnote ⁴⁶ Those oft-quoted wordsFootnote ⁴⁷ have become part of the lore surrounding our federal system.

But there are major problems with that defense of states. To begin with, even on those occasions when state experimentation has produced seemingly positive results – and I don’t deny there have been many such instances – the credit often belongs elsewhere. Charles Tyler and Heather Gerken have demonstrated that the real credit for innovation belongs not to the states, but to the third-party interest groups and funders who motivate the states and influence the federal framework.Footnote ⁴⁸

Perhaps more important, there are so many ways in which these “experiments” not only have failed to generate actual benefits, but have caused great harms. Michael Wishnie has decried “state anti-immigrant employment restrictions” as well as “state or local restrictions on immigrant access to student financial aid and free primary and secondary education.”Footnote ⁴⁹ What some euphemistically call laboratories of “innovation,” Wishnie calls “laboratories of bigotry.”Footnote ⁵⁰

Other writers, referring to the wave of antidemocratic state actions, have also preferred alternative terms for these sorts of state experiments: “laboratories of [o]ligarchy,” “laboratories of autocracy,” and “laboratories of authoritarianism.”Footnote ⁵¹ Similarly, Gowri Ramachandran & Matthew Germer have specifically criticized the spread of state actions auditing election results based on false claims of voter fraud.Footnote ⁵²

Indeed, in the very case in which Justice Brandeis uttered his famous laboratory of experimentation quote, the state “experiment” that his dissenting opinion would have sanctioned was Oklahoma’s licensing requirement for anyone who wanted to sell ice. If the Court had let this Oklahoma law stand, the practical result would have been to allow the state to deny such licenses whenever it found that that state’s one existing ice company was adequately serving the state’s residents – in other words, the creation of a monopoly for one favored company.

Even Judge Jeffrey Sutton, a staunch fan of states’ rights, chooses to use just one example to illustrate what he sees as the benefits of letting states innovate. Remarkably, the example he selects is the varying state responses to the COVID-19 pandemic. Commendably, he begins by conceding the obvious – that this example is “an unlikely candidate for illustrating how federalism works,” because “[i]t’s not a parochial problem with parochial effects; [i]t’s an existential threat to everyone, all people, all American governments. And it’s a problem that does not respect borders.”Footnote ⁵³

Precisely. But he nonetheless lauds giving the states a wide berth in responding to COVID, arguing that “it’s a problem in which borders add tools and flexibility for fixing the problem,” mainly because no one leader has all the answers.Footnote ⁵⁴ He neglects what for me stands out as the most tragic consequence of that devolution: the reckless, irresponsible policies of some states that fostered the spread of a dangerous and highly contagious disease throughout the nation. How many of the more than one million American deaths occurred needlessly because some governors and legislatures refused to regulate large public gatherings, or affirmatively discouraged vaccinations, or openly mocked those who wore masks? We shall never know. But if any issue cries out for uniform national regulation, surely it is the containment of a highly contagious fatal disease that flows like water across state lines.

Apart from the kinds of affirmative harms described above, Rubin and Feeley have provided an important insight into a problem inherent in the very concept of states innovating solutions to national problems:

Thus, the authors conclude, in a federation the different innovations don’t help us solve common national problems.

That insight reinforces my final point. Local governments, lacking sovereignty, do not present the same obstacles that states currently do. Without states, the national government could formulate common goals on those issues that are suitable for national regulation. The innovations of local governments could therefore be directed to achieving those goals, not obstructing them.

Nor is it just a matter of sovereignty. There are more local governments than there are states – far more. If 50 political subdivisions are thought to provide the machinery for social experimentation, imagine what 90,000 would do. To the extent that the argument rests on there being multiple governmental units to experiment with a variety of approaches, therefore, we don’t need states. Local governments could serve that function at least as well, especially without state governments tying their hands. Even Friedman, an enthusiastic proponent of the innovation rationale, is careful to couch his description as one of state “and local” experimentation.Footnote ⁵⁶

Sometimes, of course, the particular experiment will be too large a project for either the population or the geographic area of a single local government. In those instances, it can partner with the national government or with neighboring general or special purpose local governments – or, depending on the subject, even distant local governments. Thus, partnerships do more than exponentially expand the sheer number of possible collaborations; they can also make experiments economically or logistically more feasible.

E Encouraging Citizen Participation

Another common refrain among federalism fans similarly rests on the premise that states are closer to the people than the national government is. This argument is that federalism thereby enables and encourages citizens to become more involved in the democratic process.Footnote ⁵⁷

But neither the premise nor the conclusion is self-evident. For starters, as Rubin and Feeley observe, “there is no theoretical or empirical reason to believe” that states are closer to the people in actual practice.Footnote ⁵⁸ There is, in fact, reason to believe the opposite. As noted earlier, at least one prominent scholar, Miriam Seifter, demonstrates empirically that civil society has been a much weaker check on state laws and policies than on those of the federal government.Footnote ⁵⁹

Second, as Rubin and Feeley also point out, “[f]ederalism does not necessarily increase participation; it simply authorizes a set of specified political sub-units – states in our case – to decide for themselves how much participation is desirable. Some might choose to encourage participation but others might choose to suppress it.”Footnote ⁶⁰ Indeed. Although their 1994 article was written decades after the elimination of both poll taxes and literacy tests, the spread of state voter suppression measures discussed in Chapter 3, Section B has since reached epidemic proportions, making their observation even more salient today.

But most importantly, returning to a common theme, any argument that state government encourages citizen participation can be made a fortiori for local governments. Heather Gerken observes that, “[b]ecause states are so large, scholars who write about bringing governance closer to the people often segue into discussions of lower-level institutions.”Footnote ⁶¹

This is for good reason. For one thing, by virtue of their closer physical proximity, local officials will generally be more accessible and more easily held accountable than federal officialsFootnote ⁶² and, I would add, than state officials. In addition, all else equal, voters might be expected to be inherently more motivated to participate in local political affairs than in state or national affairs because their preferences – whether expressed in the form of a vote or any other form of political participation – are more likely to make a difference. Then too, many neighborhood citizen advisory boards offer their local governments valuable ground level input into the needs and preferences of the communities they serve.Footnote ⁶³ For all these reasons, the transfer of much of the states’ decision-making to local governments would expand the terrain over which this advantage could be realized.

Richard Briffaut has an interesting take on these issues. He readily acknowledges that “the values said to be advanced by federalism are not distinctively associated with the states. Many of these values [including political participation] may be served better by local governments than by states.” I fully agree. But he goes on to argue that that’s not the point. Preferring a formal approach to federalism over a normative approach, he maintains that characteristics like fixed boundaries and autonomy over spending (in those subject areas not reserved for the national government) leave the states better positioned than their constituent local subdivisions to discharge the functions of federalism.Footnote ⁶⁴ With that, I am not sure I agree. It is unclear, for example, how either fixed boundaries (which local governments also typically possess, for that matter) and partial spending autonomy render states better suited than local governments to encourage political participation. But even if one accepts Briffaut’s factual assumption, his argument does not identify any value that federalism serves. At most, it suggests that, given federalism, those functions that are to be performed at some subnational level – or at least some of them – are better performed by the states than by local governments. It does not explain why we need federalism, or even state government, in the first place.

F Citizens’ Voice in National Policies

A state’s congressional delegation, acting collectively, will generally have more muscle than a single member of either the House or the Senate. They can use that extra muscle to call the attention of Congress or the Administration to a problem that has arisen in their state and to push for a remedy that only the national government has the resources to provide. In that way, the argument might run, the current federal-state system strengthens the voice of every state’s citizens in the formulation of national policy, especially on matters that affect them in distinctive ways.

But that assumes that the members of the state’s congressional delegation share a common incentive to spend their political capital on that problem. That assumption, in turn, ignores the fundamental demographic reality that makes the division of our country into states so politically arbitrary to begin with. Common problems and common political preferences are extremely unlikely to map so neatly onto the geographic territories of the states. To return to a recurring theme, a given problem is much more likely to be confined to one area within the state (often mainly urban or mainly rural), or common to parts of different states, or national in scope, than to be congruent with the area defined by the frequently arbitrary borders of a single state. Thus, the members of Congress who serve the affected geographic areas (or all of Congress if the problem is nationwide) will be a more receptive coalition than a delegation of members whose only connection to one another is that their districts happen to lie in the same state or states. We are back once again to the same familiar reality that casts doubt on most of the claimed benefits of states: When it comes to serving as proxies for the varying interests or political preferences of subnational populations, the states are no match for the local governments.

G State Constitutions and Individual Rights

Judge Sutton points out that most state constitutions differ from the US Constitution in important ways. His more significant examples include popular election and retention of judges; a plural executive in which several state leaders, not just the chief executive and second-in-command, are independently elected; balanced budget requirements; initiatives and referenda; and a relatively easy constitutional amendment process.Footnote ⁶⁵ Jessica Bulman-Pozen and Miriam Seifter identify some of the same examples and add others, including equipopulous Senate districtsFootnote ⁶⁶ and explicit recognition of the right to vote.Footnote ⁶⁷ All this, they rightly point out, adds up to final constitutional products that are generally more majoritarian than their federal cousin.

The point of Bulman-Pozen’s and Seifter’s excellent article is to call attention to state constitutions’ promotion of democratic norms. Judge Sutton’s similarly thoughtful book seems aimed at a broader objective – to extoll federalism and to urge a broader role for state government in our federal system.

But whether or not so intended, in my view none of these differences between the federal and state constitutions strengthens the arguments for federalism or even for having states at all. Some of my reasons for saying this relate specifically to the individual examples; my other reasons apply more generally to all the examples.

Let’s start with the specifics. Although there are reasonable arguments to be made both for and against the popular election and periodic retention of judges, I regard the practice as a serious weakness, not a strength, in the state systems that follow it. Still, I leave that discussion to Chapter 6, Section C, as part of the outline of my proposed new republic. Whatever one’s preferred method of selecting judges, there are more general reasons, discussed below, not to view this feature as a benefit of having states.

As Bulman-Pozen and Seifter observe, the plural executive that most state constitutions require can be thought of as majoritarian in nature. But it is majoritarian only in the same sense that initiatives, referenda, and even popular selection of judges are majoritarian: all these measures exemplify direct democracy. And direct democracy is not the only form, or in the United States even the primary form, of democracy. In our representative democracy, the people delegate to their elected leaders the power not only to pass laws and make policy decisions, but also to choose other policymaking officials. Which of these officials should be elected, and which ones should be appointed by others who have been elected, is a question of degree and therefore a matter of opinion.

A constitutionally entrenched balanced budget requirement, in contrast, is not a majoritarian feature. To the contrary, it is explicitly counter-majoritarian, since it prohibits the people’s elected representatives from enacting a budget deficit when they deem it beneficial to do so – for example, by cutting taxes or authorizing stimulus spending in times of recession. Economists will forever differ over the relative pros and cons of long-term budget deficits, but the point here is that the prohibitions contained in various state constitutions are not a majoritarian example.

Popular initiatives and referenda similarly have both pros and cons, but I agree with Bulman-Pozen and Seifter that they can be fairly characterized as majoritarian instruments. Still, they merely substitute direct democracy for representative democracy, and at any rate, the discussion in Chapter 3, Section F describes the modern trend of state governments overriding or preempting popular initiatives. Moreover, as discussed below, local governments also frequently allow popular initiatives and referenda. Under the model proposed in this book, they would remain able to do so, and national initiatives and referenda would be permitted as well.

The relative ease with which state constitutions can be amended is clearly a majoritarian feature, especially when accomplished by a direct vote of the people. But constitutions protect fundamental individual rights and essential institutions of government; they are not meant to be too casually altered. Some measure of durability, even though inherently counter-majoritarian, is important. While I argued in Chapter 2, Section E that the US Constitution tilts too heavily toward durability, my view is that the power to change state constitutions by simple majority vote tilts too heavily in the opposite direction. While I therefore don’t see this feature of many state constitutions as an argument for keeping states, I acknowledge that those who favor an easy or more purely majoritarian constitutional amendment process might well feel otherwise.

Bulman-Pozen and Seifter point out that state Senates, unlike the United States Senate, are equipopulous. And they rightly regard that component as a prerequisite to political equality, which in turn is a basic pillar of a true democracy. Judge Sutton, tellingly, does not include that feature in his list of arguments for federalism or for states; to do so would have required an acknowledgment that the contrary composition of the United States Senate, necessitated by the notion of equal state sovereignty, is undemocratic. But before giving the states credit for adopting greater legislative majoritarianism than the federal government does, let us remember that it was the US Supreme Court, in Reynolds v. Sims,Footnote ⁶⁸ that forced them to do so. At any rate, without states, the US Senate could, and under my proposal would, become similarly equipopulous, as discussed in Chapter 6, Section B.2.

As noted, many state constitutions commendably enshrine the right to vote. The US Constitution does not. It merely specifies certain impermissible forms of discrimination in voting.Footnote ⁶⁹ Its only recognition of a right to vote is the Seventeenth Amendment requirement that US senators be “elected by the people” of each state.Footnote ⁷⁰ My proposal envisions explicit recognition of a right to vote – directly for the president upon abolition of the Electoral College, and by a separate, similar provision for election of members of the Senate.Footnote ⁷¹

Apart from these provision-specific responses, there are at least three general reasons that none of these distinctive characteristics of state constitutions is a persuasive argument for federalism or even for states. First, in the absence of state governments and the constraints they currently impose on their constituent local governments, the latter could decide on their governmental structures themselves. If the residents of one municipality wanted to choose their mayor and city or town council by direct election and let the mayor appoint all the other executive officials, while another municipality preferred direct election of the city or town manager, the city attorney, and any other officials, both of those municipalities could do as they wished. The same is true of the decisions whether to require balanced budgets, or permit legislation by popular initiative, or make it relatively easy to amend the city or town charter, or enshrine the right to vote in the local government charter, or elect the city or town or county legislative council through equipopulous districts, or provide for popular election of municipal judges.Footnote ⁷² If democratic norms call for allowing each state to choose its own governmental structure, then they call even more loudly for local cities, towns, and counties to have the right to do so. Even Judge Sutton ends a chapter with the following conclusion: “Democracy indeed may work most effectively at the local level today.”Footnote ⁷³

Second, whatever one’s view of the merits of each of those issues, it’s hard to see how those merits vary depending on the needs of the particular state. At either the national level or the state level, do elected judges and elected attorneys general and equipopulous Senate districts make more sense for residents of Michigan than for residents of Alabama? Would the availability of popular initiatives, or easier amendments of a constitution, be better for folks in South Carolina than for those living in Oregon? Are there states where people would be better off if the right to vote were not recognized? Offhand, I can’t think of anything one might find in the Wisconsin waters that is lacking in the Colorado Rockies that makes it especially important to elect judges and officials rather than let the chief executive appoint them, or to prescribe a plural executive, or to require balanced budgets, or to allow legislation by popular initiative, or to ease the path to constitutional amendments.

Some might argue that that is not the point. Even if the merits of the competing arguments don’t depend on the distinctive needs of the particular state, the decisions should depend on the political preferences of that state’s citizenry. That is, after all, the essence of democracy.

Whatever force that argument is thought to have, it is not a reason to retain state government, mainly because the needs and preferences of a given state are rarely either internally uniform or externally distinctive. Again, the prevailing political preferences of Atlanta’s citizens bear little resemblance to those living in rural Georgia. When it comes to political polarization, the urban-rural divide is far sharper than the divides among states.

Third, state constitutions are only as good as the interpretations that the states’ Supreme Court justices place on them. Who are those justices? In many states, judges are elected (and retained or terminated) by the same voters who choose whether to elect and reelect the state legislators. And the vast majority of states are safe states – that is, reliably red or reliably blue. So one can reasonably assume that in most states the political views of the judges will generally not stray far from those of the legislative majority. That matters. As discussed in detail in Chapter 2, Section D, interpretation is often – and in the case of constitutions, usually – far from cut and dried. It requires judgment, and judgment is unavoidably affected by one’s ideological – or, dare I say, partisan – inclinations. State constitutions can certainly be valuable tools. But for the task of taming counter-majoritarian state action, they are a thin reed on which to rely. Bulman-Pozen and Seifter cite decisions by the Supreme Courts of Pennsylvania and North Carolina striking down legislatively gerrymandered maps as violations of their respective state constitutions,Footnote ⁷⁴ but it was not long afterward that a partisan shift in the North Carolina Supreme Court’s majority prompted that court to overrule its earlier decision with breathtaking speed.Footnote ⁷⁵

H Personal or Political Community

Some have argued that federalism provides a sense of community. As Rubin and Feeley explain, two main forms of community have been invoked in service of this argument. There are what some have called “affective” communities, in which “members feel a personal or emotional connection with one another,” and there are “political” communities, in which members “engage in a collective decision-making process.”Footnote ⁷⁶

But both strands of the community argument assume that a state’s residents share some important commonality and that that commonality distinguishes them from the residents of other states. Rubin and Feeley debunk that assumption as well as anyone has. They point out that “affective communities necessarily consist of small groups.”Footnote ⁷⁷ States are not nearly small enough to serve that purpose. As for political communities, “Most of our states, the alleged political communities that federalism would preserve, are mere administrative units, rectangular swatches of the prairie with nothing but their legal definitions to distinguish them from one another.”Footnote ⁷⁸

Recall that the federalism argument discussed in Section B relied on the claimed ability of states to tailor their laws and policies to the distinctive needs and preferences of their citizens. The federalism defense considered in this section relates to that same theme but is slightly different. Where the earlier defense focused on supposedly common interests and therefore preferences for particular laws and policies, this one focuses on community as a personal or political end in itself. Citizens of a given state are depicted as having characteristics that leave them bound to one another in ways that do not bind them to others. This, the argument runs, enables them to forge bonds, and pursue community interests, that are not realistic for the much larger national population.

While the arguments considered in these two sections of the chapter differ slightly, they contain the same fatal flaws: the sheer sizes of states and the crudity of their boundaries as a proxy for delineating these groups. As Section B demonstrated, the most dramatic divides are not interstate, but intrastate. Let us assume for the sake of argument that states, presumably because of their smaller populations and geographic areas than the nation as a whole, are more likely than the federal government to foster – and protect – a sense of both political and personal community. If that is so, then the same attributes – smaller land masses and smaller populations – would suggest that local governments are still better suited to that purpose than states.

I States as Federal Partners

Some scholars have touted, or at least acknowledged, the advantages of states partnering with the federal government to implement federal programs.Footnote ⁷⁹ Bowman et al. identify states’ closeness to the people as a reason that they can adapt public programs to local needs in ways that are administratively efficient.Footnote ⁸⁰

Of course, if closeness to the people is what makes states good partners in implementing federal programs, then local governments, being closer still, should make even better federal partners. Many of the concrete examples offered in Chapter 4, Section C.3 illustrate the value of national-local partnerships in discharging functions that would be reassigned to the local governments if state government were abolished. In addition, regional arrangements of local governments have an advantage as partners that states don’t have: they can cross state lines. Those collaborations can be voluntary or, as Nestor Davidson has noted, accomplished with a nudge from Congress – as has been done in the past.Footnote ⁸¹

Still, one might assume, it is a lot easier for the national government to deal with 50 state partners than to deal with 90,000 local partners. Yet, they do it all the time. Davidson refutes the conventional view that the only direct relationships between two levels of government are federal-state and state-local. To the contrary, he shows,

Moreover, as noted in Chapter 4, Section B, the federal government can, and frequently does, operate through regional offices that interact with all the states in the region. In the absence of states, it could similarly interact with all the local governments in their respective regions. Additionally, there would be no need to provide each regional office with its own separate legislature, its own separate governor’s office, and its own separate judiciary, attributes required for the states whom they now use as intermediaries.

These sorts of direct national-local relationships would be easier still in a unitary republic. There would be no state sovereignty and therefore no anti-commandeering doctrine. As discussed earlier, in the anti-commandeering cases the Supreme Court has prohibited the federal government from requiring states to help implement federal programs. Unless every state consents – not always the case in our increasingly polarized country – it can be virtually impossible to operate a nationwide program that relies on local implementation. In our current federal system that prohibition is absolute – no balancing of competing interests is involved – and is not limited to the law enforcement context.Footnote ⁸³

Without state government, there would also be no assertions of plenary state authority over local governments. In Davidson’s view, local governments should not be held subservient to the states in which they are located. Rather, he argues, the national government can be both the source of, and the constraint on, local governments’ autonomy.Footnote ⁸⁴ I very much agree. But while Davidson’s point is that the Supreme Court should embrace that view (rather than a more sweeping version of state sovereignty) as a matter of constitutional interpretation, I am suggesting, for all the reasons included in this book, that the US would do well to go two steps further. I would eliminate not just the concept of plenary state power over local governments, and not even just state sovereignty, but the entire institution of state government.

Jessica Bulman-Pozen argues that, when Congress enlists states as partners in the implementation of a federal program, the states play a useful separation of powers role. They do this, she maintains, by challenging federal executive branch actions that the states believe exceed the executive’s statutory authority.Footnote ⁸⁵ It is an interesting thought, and I don’t quarrel with it. For present purposes, it is enough to say that in the unitary American republic considered in this book, local governments would be equally well positioned to perform the function of challenging executive actions on those sorts of ultra vires grounds. In many cases, private parties who are adversely affected by the national executive branch actions would similarly be able to bring the case to court.

J Method to the Madness: Boundary Locations

When this book was in its early stages, its thesis drew more than its share of skeptics. Most likely, it still does. One friend asked me (and I’m paraphrasing) “Why were the various state boundaries drawn where they were in the first place? Surely there must have been some rhyme or reason to the locations of those lines. And wouldn’t the reasons for separating the states along those specific boundaries reveal reasons for having states in the first place?”Footnote ⁸⁶

It was an intriguing question. In searching for the answer, I stumbled upon an exceptionally helpful article by John Harrington & Grant Suneson.Footnote ⁸⁷ The authors carefully researched and described the origins of each state and its boundaries. From the collection of those individual descriptions, I take the liberty of breaking the inquiry into two parts: How did each state become a state? And why were its specific boundary lines drawn where they were?

The technical legal answers to those two questions turn out to be the same: “it’s a mishmash.” As to the “how” question: The thirteen original states, of course, were formed out of the British colonies and became US states by ratifying the new Constitution. Some of the other states had been US territories or carved out of US territories. Still other states resulted from the partitions of existing states. Some states were admitted in pairs, one slave state and one free state. And some states (e.g., California and Texas) had been independent countries before being admitted to the Union.Footnote ⁸⁸

As to the “why” question: The explanations for the fifty states’ precise boundaries are similarly diverse. Most of the thirteen original states simply retained their colonial boundaries. And most of those boundaries had been laid out in charters, deeds, or patents granted by the King, often in keeping with natural geographic barriers. Those natural barriers – seacoasts, gulfs, bays, lakes, rivers, mountains, and in the case of the Dakotas the edge of a plateau – would likewise determine the boundary lines of many of the subsequently admitted states. As the nation developed, canals and railroads began to surpass rivers as the primary means of both trade and travel; increasingly, they too became state boundaries. Stillother state boundaries were fixed by either latitude lines (“parallels”) or longitude lines (meridians). Some of those were set so as to accommodate President Jefferson’s predilection for states of similar size and shape, often a recipe for otherwise arbitrary boundaries.Footnote ⁸⁹

At least with respect to the drawing of the boundaries, this patchwork might be analogized to the European colonization of Africa and the ultimate division of that continent into its current independent nations. One scholar, recounting the history of African colonization, could as easily have been describing the origins of the American states when she said: “Lines of longitude and latitude, rivers and mountain ranges were pressed into service as borders separating the colonies. Or one simply placed a ruler on the map and drew a straight line.”Footnote ⁹⁰

For present purposes, the key takeaway is that none of those determinants – royal charters, natural geographic barriers, canals and railroads, latitude and longitude lines, or a desire for states of equal geographic areas or similar shapes – had anything to do with the ethnicities, religions, interests, or political preferences of the state’s inhabitants.

There are a handful of arguable exceptions. The populations of the slave states and the free states had sharply differing interests, and their admissions in pairs reflected political compromises. But those differences don’t explain the precise locations of their boundary lines. Similarly, at least two early border disputes – Pennsylvania/Maryland and Virginia/West Virginia – were settled by separating the states along the Mason–Dixon line. California’s eastern boundary could not be pushed further east because the Mormons were in the process of establishing a separate territory. And Rhode Island was shaped by Roger Williams’s desire for a colony based on religious tolerance.Footnote ⁹¹

But even if those few examples are seen as exceptions to the demographic randomness of the state boundaries, they are pretty much the only exceptions. And even if most of the state lines had been drawn to accommodate demographic differences among the nation’s people, demographics change over time – and with increased speed as technology makes interstate migration ever easier. For that reason too, the demographic differences are too transient to explain the need to divide the nation into fifty permanent states. Moreover, as the discussion in Section B makes clear, it is the local governments, not the states, that provide the better proxies for the differing interests and preferences of the nation’s subnational populations.