State Bankruptcy and the Federal Order

While a federal bankruptcy law for states might be a desirable policy, its constitutionality is doubtful. Now, I am not a constitutional lawyer and cannot speak to the details of United States case law, but I do study fiscal federalism in comparative context: its conditions, operations, and consequences. Accordingly, I will analyze how a federal law for state bankruptcy would tend either to support or to undermine the values one might seek to protect by means of the federal, constitutional order in the United States.

The main points of the argument are as follows. First, federal bankruptcy law for cities reaches around states’ traditional, sovereign authority over local government, and the New Deal Court’s reasoning on this point might not have been sound. Second, federal bankruptcy protections for state governments risks a kind of federal-state collusion that undermines the true purposes of federalism, which are to protect private citizens, not the governments comprising the system. Third, many states went bankrupt in the 19th century, and they have their own legal remedies available to reduce unpayable debts, making a federal law unnecessary.

On the first point, while it is true that municipalities are not mere arms of state government in the way that counties typically are, they owe their existence to state government. They are incorporated under state law. So are private corporations, but a difference from private corporations is that state legislatures may alter or abolish municipalities as they see fit. While private incorporation looks more like state registration of an entity created by private actors, municipal incorporation looks more like state creation of an entity, though typically by petition of private actors.

The degree of state legislative discretion over municipal creation, abolition, and merger is much greater than over private corporations. Certainly, some states have afforded municipalities “home rule,” but even this status is something that states have granted on their own discretion, not a federal constitutional right that municipalities can claim against interference from state governments (including state constitutional amendments).

Municipalities are governments, bound by federal and state constitutions. They are distinct from homeowners’ associations, which are truly private providers of local services, and that distinction makes a real difference: for instance, homeowners’ associations are not bound to implement universal, equal suffrage. For the purposes of most constitutional law, municipalities are considered governments, not private entities. The exception for federal bankruptcy law looks ad hoc.

While I again concede my lack of expertise in constitutional law, it is difficult for me to see how the status of cities vis-à-vis states and the federal government differs much from the status of the Bank of the United States vis-à-vis states and the federal government during the 19th century. The Bank of the United States was a creation of the federal government, though not an arm of it. Nevertheless, the Court ruled in McCulloch v. Maryland (1819) that states could not impede its operation. If states may not detract from federal sovereignty over institutions of federal design (within the scope of Congress’ enumerated powers), how is it possible that the federal government may limit states’ authority over institutions of states’ design (within the scope of constitutionally recognized state sovereign powers)?

United States v. Bekins (1938) may then have been wrongly decided. By extending federal bankruptcy procedures to cities, even with the consent of states, Congress took away from states some of the burden of responsibility they shoulder for the local governments they create and supervise. The Constitution does not prohibit states from enacting bankruptcy procedures for themselves or for the public institutions they create and supervise.

The second worry about extending federal bankruptcy protections to states comes up in Michael McConnell’s Liberty Forum essay, but it’s worth amplifying: Federal bankruptcy protections for states would protect governmental entities from fulfilling their obligations to private actors. That looks like federal-state collusion to deprive citizens of what they are owed. The purpose of a federal system is not to protect the rights of state governments, but to protect the rights of citizens.

In the Founding-era debates over the Articles of Confederation and the 1787 Constitution, concerns about liberty come up repeatedly. Anti-Federalists feared what a consolidated, general government would do to liberty, and they thought the Constitution created a government that would eventually devolve in that direction. Federalists feared the depredations of state governments, which they were living through in Rhode Island, Massachusetts, and elsewhere (“a rage for paper money, for an abolition of debts, for an equal division of property,” as Madison put it in Federalist 10). The way both sides conceived the matter is that a well-designed federal system would allow federal and state governments to check each other and preserve individual liberty.

This way of thinking about federalism wasn’t quite right. While the federal government certainly can check states, through congressional legislation that overrides state legislation via the Supremacy Clause or through federal appellate court decisions, states cannot reciprocate by vetoing (“nullifying”) federal laws. Another flaw is that while the Federalists believed that the Senate—whose members were, in the original constitutional design, elected by state legislatures—would fulfill this “checking” function, it turned out that state legislatures are often willing to relieve themselves of the burden of autonomy while taking compensation in the coin of federal transfers.

Incidentally, lest anyone think the Seventeenth Amendment was responsible for the decline of competitive federalism in the United States, consider the example of the Federal Republic of Germany, where the upper house or Bundesrat is still elected by Land governments, but where an initially competitive system of federalism was abolished in the 1960s in favor of a collusive model that subsidizes nearly all of each Land’s budget through harmonized, federally determined taxation.

Better protections of individual liberty in the United States are to be found in three other mechanisms: 1) democratic control, 2) sorting, and 3) competition.

Democratic control of government might be greater for smaller-scale governments, where an individual’s vote is more decisive and representatives have incentives to be more attentive to individual constituents’ concerns.

Sorting lets citizens find jurisdictions to live in where the laws are more to their liking, thus reducing the number of Americans living under legal regimes they find harmful or inconvenient. Sorting allows for more policy diversity across jurisdictions.

Closely related to sorting, competition encourages states to reduce their extractions from private citizens because such extractions will reduce the tax base and therefore revenue. Competition forces less policy diversity across jurisdictions when it comes to those policies that few private citizens would reasonably want, such as rent-seeking.

How would a federal bankruptcy law for states affect these functions of fiscal federalism? It might have salutary, first-order effects on competition: letting states go bankrupt would raise borrowing costs for all states, as creditors became less willing to lend, encouraging states to borrow less and tighten their budges. If states are currently excessively spendthrift, a federal bankruptcy law is likely to curtail this vice somewhat.

Yet it deserves to be mentioned that not all states are spendthrift. Eileen Norcross and Olivia Gonzalez of the Mercatus Center produce an annual study of states’ fiscal conditions. Their top state, Alaska, has used oil revenues to build a massive net asset position:

“[T]he state has between 22.46 and 23.44 times the cash needed to cover short-term liabilities. Revenues exceed expenses by 55 percent, producing a surplus of $8,296 per capita. On a long-run basis, net assets represent 85 percent of total assets, and liabilities are 3 percent of total assets.”

Similarly, states like Nebraska, Wyoming, and the Dakotas have built up large rainy day funds while keeping debt and pension obligations low. It might be unfair and unwise to allow these states’ interest payments to rise as a result of the low-probability risk introduced by the possibility of bankruptcy. Residents of these states might benefit from additional public spending to provide collective goods, even when it is debt-financed. It is possible to constrain state debt too far.

Bankruptcy protections for states might therefore interfere with the democratic control and sorting features of fiscal federalism. The preferences of citizens of different states might differ when it comes to public spending and taxation. A uniform bankruptcy law for states could reduce citizens’ ability to implement these preferences into law. If the citizens of Illinois want to require their children to pay a large share of their income for public employee pensions, the federal government perhaps should not allow them the “easy out” of bankruptcy—which would in fact not be so easy ex ante, since Illinois would have to pay higher interest on its debt as a result of the possibility of bankruptcy.

The more significant worry about a federal bankruptcy law for states is second-order. A uniform law of bankruptcy would be a target for state lobbying and political bargaining. States that might want to go into debt in the future (say, for large infrastructure projects) would press for a strict bankruptcy law to keep interest rates low, while those that are already approaching insolvency would presumably push for the loosest possible terms. Creditors, pensioners, and companies with large state tax liabilities would all have incentives to lobby for their preferred solutions. Once it became possible for federal lawmakers to play favorites among state governments through U.S. bankruptcy law, Congress would undermine the constitutional autonomy of the states.

That brings us to the final point: A uniform, federal solution for insolvent states is unnecessary. When states can no longer pay back their obligations, they can do what insolvent states in the 1840s did: default on or repudiate their debts or pension promises. To the extent that state constitutions impede these solutions, there is always the amendment process. If voters refuse to amend the constitution of their insolvent state, they will have to feel the fiscal squeeze until it becomes obvious to everyone concerned that a drastic solution is inevitable. This messy and difficult learning process is occasionally necessary for democratic federalism to work.

Unlike Puerto Rico, states are sovereign. The federal government cannot force states to pay back their debts. Default and repudiation do not violate the federal Constitution. When these actions violate state constitutions, each state’s own supreme court is the final court of appeal on legal disputes arising from those actions.

Letting insolvent states default on their own debts according to their own legal and political processes has several virtues. First, it allows the more profligate states to suffer the consequences of their own decisions. Creditors already factor in the possibility of state default. Standard and Poor’s rates Illinois’ general obligation bonds BBB+ while South Dakota is AAA. As a result, Illinois pays higher interest rates on state bonds than does South Dakota. Furthermore, these ratings reflect each state’s creditworthiness, namely, the risk that bondholders will not be paid back in full and on time. Illinois has more state debt and unfunded pension obligations as a share of its private economy (the tax base) than does South Dakota. The interest rates that each state faces are thus rationally related to those states’ different policies.

In his 2006 book Hamilton’s Paradox: The Promise and Peril of Fiscal Federalism, Stanford political scientist Jonathan Rodden finds that in the United States, state debt burdens are strongly negatively correlated with their credit ratings. This is good evidence that the market thinks that the federal government will not bail out bankrupt states. States that default will have to pay the costs of their own poor decisions—in higher interest payments and tax burdens—rather than externalizing those costs onto taxpayers in other states. This is a desirable arrangement because it establishes the appropriate set of incentives for state governments to maintain fiscal discipline and make optimal choices about whether or not to extend state debt.

The second advantage of letting states default on their own terms is that in the regime of reasonably hard budget constraints described in the preceding paragraph, remaining differences across states in debt burdens and credit ratings reflect either differing citizen preferences across states, or differing qualities of state institutions. In either case, permitting variation across states is desirable.

To expand on an argument already given, if citizen preferences differ across states, we can satisfy more citizens by allowing those preferences to be translated into policy. Perhaps the median voter in Illinois is more risk-acceptant or discounts the future more than the median voter in South Dakota. Requiring the citizens of Illinois to behave more like the citizens of South Dakota, as a federal bankruptcy law might do, would then make Illini worse off according to their own (perhaps not-so-virtuous) preferences for profligate living.

Plausibly, the differences between states’ fiscal policies reflect not just citizens’ tastes, but also differences in institutional quality—differences that provide important information to voters and policymakers everywhere. Perhaps Illinois’ state institutions are lower-quality than South Dakota’s (public employees have excessive collective bargaining protections, or legislators are less attentive to local median voters, or the lawmakers in Springfield are simply more corrupt than those in Pierre). In that case, Illinois ought to change its institutions to be more like South Dakota’s, and they in fact have incentives to do so. But in the absence of different credit experiences, the citizens of Illinois might never have come to realize that their institutions yielded worse outcomes than those of South Dakota.

In conclusion, a uniform bankruptcy law for states might be desirable if it were evenhandedly designed. However, it might not be constitutional, and it would certainly run against the values that a proper fiscal federalism can secure. Moreover, we cannot count on a federal bankruptcy law for states to be evenhanded and well-designed, and in any case, insolvent states can always default on or repudiate their respective debts even without a federal law authorizing them to do so.