Some commentators think that the two main victims of bankruptcy by a state government would be unionized state employees and state bondholders.
I'm not sure about that result. In practice, when national governments get in debt over their heads, they tend to either dishonor their obligations (e.g. Confederate bonds and bonds from Iceland), sometimes only temporarily while issuing IOUs (a la California's recent experience), or enact austerity plans that increase taxes and reduce public spending together with a restructuring of the timing of debt obligations (as Ireland and Greece recently did). But, courts have also been deeply reluctant to order tax increases, even when a state is not meeting its judicially determined obligations to pay its debts. It isn't even clear if this could ever be consistent with the Republican government clause of the United States Constitution.
Given creditors have few involuntary collection remedies against state governments, it is perhaps remarkable that they pay as reliably as they do.
Public employees are typically paid in a timely manner, and have their pension funds deposited in an account in which they arguably have property rights that cannot be abrogated by legislative act (although others would argue that the defined benefit and not the funds set aside for payment of defined benefits are the things in which public employees have pension rights, contrary to recent Colorado action reducing future defined benefits). The pensions are often underfunded, but in practice, determining funding sufficiency is more art than science, and I am not aware of any state pension that is so deeply underfunded that it cannot meet its current payout obligations for the next few years. Underfunding is typically discussed in terms of the ability of pension funds to meet their obligations a decade or more in the future given conservative actuarial and investment return assumptions.
Unionized employees also have union contracts that limit flexibility in adjusting their pay, but typically not preventing layoffs. But, it isn't obvious that the government can hire enough qualified employees without offering something close to what it does to its employees. Moreover, if bankruptcy reorganizations offer any lesson, it is that trade creditors (i.e. those necessary for the ongoing day to day conduct of the business) usually come out better than their formal legal preference in the process would suggest, because their refusal to offer further services would typically make a reorganization impossible.
Also, while American public sector labor relations have historically been rather placid, there are numerous international examples of public employees exercising immense practical power through labor union coordinated labor actions (whose legality is often irrelevant for practical purposes, unless one is willing to fire them all as Reagan did in the aircraft controller's strike). An unprecedented deep blow to unionized public employees via a state's bankruptcy, could be just the thing to trigger that latent power. Even a credible threat of a European or Latin American class public employees strike might give public employees considerable bargaining power in a bankruptcy negotiation.
Putting the squeeze on public employees via a state bankruptcy is also complicated by the fact that many public employees either have deeper constitutional protections from pay cuts (e.g. judges), a compensated out of a dedicated funding source (like Secretary of State filing fees) that has not been exhausted, or are a step removed from state government, with direct employers who are local governments (e.g. school districts) or autonomous entities (like public universities and airports and hospitals) that may themselves not be bankrupt and unable to meet their obligations, even if the state that has sponsored them is unable to pay its bills. A surprisingly small share of all public employees are compensated directly out of state general funds by state government, and some of them (e.g. prison guards and tax collectors) are difficult to fire en masse in a revenue enhancing, public safety tolerable way.
Vendors to the state, such as doctors owed Medicaid reimbursements and non-profits that have been awarded grants to provide some public service, might be in a weaker position than public employees. They may cease to do business with state government to cut their losses, and the state may be able to tolerate significant levels of defection of that kind.
Municipal bondholders might see their payments deferred significantly, with interest at the municipal bond rate, but not discharged. They might, in other words, be allowed to enter judgments against the state, but not allowed to enforce those judgments. This might make it impossible for the state to borrow new debt at less than crushing subprime interest rates, or without some special form of collateral, guarantee or creditor protection.