Almost all government bodies issue at least some debt in this fashion, as far as I am aware, and I don't know of one that would be stupid enough to try to back out when the debt comes due. Sometimes they go for this when the issuance rules for general obligation bonds are too strict (like requiring voter approval) or because it allows them to keep their official debt burden low as a percentage of their taxable base. I consider the practice both dishonest and a great candidate for a national reform movement, but RI isn't doing anything particularly unusual (the dishonesty is mitigated by the fact that all of the numbers are widely available; it's a practice designed to get around inconvenient self-imposed laws rather than outright fraud).

Nevertheless, no government would ever back out. If they did, they would simply never be able to sell bonds at a decent rate again. It would mean, for instance, that state workers could wait months between paychecks because the state wouldn't be able to use TANs. The cities and towns would all suffer as well -- they enjoy the same sort of implicit state guarantee on their debt. In fact, if a state did default, I'd be willing to bet that the Federal government would step in to try to solve the problem, as the states enjoy an implicit guarantee from them just like Fannie Mae did (this wouldn't necessarily go for large states, like California, that may be too big to save, and would undoubtedly come with new costs so that the debtor state eventually repaid the rest).

It's true that there is no legal obligation to repay the bonds, but this is as forceful as "no legal obligation" gets. Anyone trying to argue that this isn't the same as GO debt is just lying.

Mario,

Two quick points, admitting fully that I had never heard the term "moral obligation bonds" before this week:

1. Much of what you say makes sense to me, if you are talking about using moral obligation bonds to back a loan, rather than a loan guarantee, the idea being that in the case of a loan for a capital project, there is supposed to be a future income stream, part of which gets dedicated to paying back the bondholders. But when you are using the bonds to back a loan guarantee, you are setting up a situation where you are promising to pay out big money only in the case where no dedicated future income stream ever materializes. This would seem to be a very questionable fiscal design.

2. When you say...

Almost all government bodies issue at least some debt in this fashion...

To me at least, the stamp of "invented by New York City in the late 1960s" isn't a ringing endorsement of unquestionable fiscal sensibility.

These things creeped me out the first time I saw them too, and I haven't really gotten over it.

I don't think I understand the distinction you make between loans and loan guarantees. The only time I have ever seen these used is when there is a separate entity involved in the deal (although most of the time quasi-public) so in a sense they are all guarantees. If it is just a municipality selling debt, it would have to be backed by the full faith and credit of the debtor. A lot of the time they are used for capital projects, but they are capital projects with a valid business structure, like a parking garage, which is why they are made into quasi-public agencies. The revenue streams for most of these projects aren't guaranteed; if the parking garaged is under-utilized, then even if it makes some money the rest will have to come from somewhere else. In this way the 38 Studios deal really isn't very different -- the dedicated revenue stream is the firm's projected profit and the public purpose is job creation. It's a little more creative than most, but you don't always have 12% unemployment. People get creative when they get desperate. A good question is whether the state would be considered a creditor if the business fails, so that the state could at least walk away with some intellectual property in the worst-case scenario. If that's the case, it could actually be a slightly better deal than most, since a failed public works-type project is more of a liability than an asset.

I should have pointed out before that one reason that these have become so widespread is that GO bonds are tax free, so the IRS is very strict about how they are used. You simply can't use tax-free bonds to finance a project with a private component, so with a mixed-use facility, like if you wanted to rent out space to a coffee shop in a publicly owned building like the DMV, you would have to account for what percentage of every pylon is holding up the private area and fund that differently or fund the whole thing using commercial paper and pay a higher than necessary interest rate. Using conduit debt allows you to maximize your savings, finance the project consistently (for instance, you can start the project before you line up private partners), and make much better use of public facilities -- so you don't have to choose between a state-owned coffee shop or none at all.

I'm not saying it's a great system, and I'm definitely not saying this is a great deal, but when used properly conduit debt makes government work better. My ideal solution would be mandate that all debt must be explicitly backed by the taxing power and just have the municipality pay a fine to the IRS to make up for benefits stemming from the private use of public credit.