Can bond markets differentiate the debts of sub-national entities?

The RBI's just released report on India's state government finances which point to the increase in state indebtedness has triggered a debate on the issue. There are several things going on. One that caught my attention about the debate that it triggered is the point about lack of differentiation of state borrowings despite the wide variations in the their fiscal balances. 

Consider two commentaries. Neelkanth Mishra, writing before the report was released, acknowledges the problem and concedes that addressing it is trickier,

The bond yields of states with low debt to GDP like Chhattisgarh and Odisha are no different from those of states with high debt to GDP like West Bengal and Punjab. This has been attributed to a range of factors : That the market assumes a sovereign guarantee (from the Union Government); that bond holders effectively has first access to state government revenues as the RBI, which co-ordinates the auctions, have access to state revenues in an escrow; state governments must request the Union government for extra borrowing beyond the prescribed fiscal deficit limit, which reduces default probability significantly; mandatory buying by financial institutions like banks generate sufficient demand; and lastly that these were too small to worry the markets about differentiation.

The Economist writes about the worrying mispricing of state government bond yields,

Which Indian state sounds more likely to repay a loan: Bihar, the country’s poorest, with a budget deficit of nearly 6% of its state GDP last year and a hole in its finances after it banned alcohol sales; or Gujarat, a relatively prosperous coastal region with a deficit nearer to 2%? According to bond markets at least, both are equally good credits, and so pay the same interest rate... Investors are lending money to all the states at the same rate for good reason: the central government, through much nudging, winking and head-shaking, has indicated it will look after them if the states default. It also compels publicly owned banks, pension funds and insurance companies to pile in.

There is no denying that the lack of differentiation is a problem. But I am not sure whether there is a solution to this in a tight federal arrangement like in India where implicit Union Government guarantee, especially when involving State governments, is strongly baked in and may be difficult to unhinge. Even with deeper and broader bond markets, this is unlikely to change meaningfully enough to allow fair price differentiation. 

Consider the example of Europe where sovereign bond yields converged spectacularly in the lead up to the monetary union and stayed there till the recent crisis. This was despite the far higher diversity, by order of magnitude, in debt to GDP ratios between Italy and Germany than between Bihar and Gujarat (or any other pair), as well as the absence of any form of fiscal union between the countries that would have facilitated transfers to support bailouts. And you cannot argue that bond markets in Europe suffered from financial repression. 

In fact, business as usual appears to be returning as markets stabilize and bond yields again start to coverage. 

It will be interesting to see the reliability of price differentiation among bond yields of state governments in the US, where the market expectations of a federal bailout can be expected to be lower. Or in other federal economies where states or provinces borrow directly from the market without any explicit federal guarantees and how the bond markets differentiate such borrowings of different states. 

I could be wrong, but apart from the US or in one or two other places due to circumstances which are sui generis and which cannot be replicated, I will be surprised if bond markets can differentiate in a meaningful enough way among the borrowings of sub-national entities in a fiscal union.