Four policy analysts at the Bipartisan Policy Center have published an eye-opening set of slides regarding how the debt ceiling blackmail crisis is actually likely to play out. It's available here, and very well worth reading.
The authors ask the following 3 questions, and offer, in the main, the following answers:
(1) What is the "X Date" - that is, the date on which the Treasury will lack sufficient cash to pay all of its bills in full and on time? Although in 2011, "extraordinary measures" bought the Treasury a full 2-1/2 months, this time around it's going to be worse. February is a less fruitful time than the summer for these games, and fewer measures are available this time. Check the slides for the full gory details, which are rich in particular institutional detail. But the bottom line is: the X date is likely to fall between February 15 and March 1.
(2) Once the X Date is reached, what would be the effect on government operations? The answer here is very grim. Consider the approach everyone has been assuming, which is to pay some bills but not others. It's "[u]nclear if [this] is feasible, given the design of Treasury's computer systems." Massive chaos is likely to result. In short, we can sit around a room and say what types of things we would prioritize if we were running the Treasury But actually implementing it might be extremely difficult, even leaving aside the massive political and legal uproar. And think of all the uncertainty for people who are trying to pay their daily bills and have money due from the government. I myself am wondering whether I should file my tax return early, in the event that I might end up with a refund. But suppose I needed a government check in order to pay next month's rent, or else to pay my employees.
"Roughly 40% of the funds owed for the month would go unpaid." This is one reason why Bernanke anticipates rapid and severe macroeconomic disaster once the X Date is reached - without even getting into questions of how our credit rating is affected.
In addition, daily "inflows and outflows do not match up well and are quite 'lumpy.'" For example, on February 15 they estimate that the government will receive $9 billion and owe $52 billion. By contrast, on February 19 it will get $15 billion and owe $16 billion.
This is one reason why I myself consider it quite plausible that the government will end up following the second payment alternative that they suggest, in lieu of prioritization. Here, rather than deciding which bills to pay - which would require the Treasury to "sort and choose from well over 100 million monthly payments" - it would wait to pay each day's bills in full until there was enough cash for the whole thing. Thus, perhaps the bills for X Day + 1 would be paid 2 days late, those for X Day + 2 would be 4 days late, etcetera. (Perhaps, if the Treasury's computers are amenable, they would pay bond creditors daily and do this for everything else.)
(3) What would be the market risks? The 2011 blackmail crisis was relatively cheap, although it will cost us $18.9 billion of extra interest payments over the ten-year period. That equals the positive 2012 revenue estimate for the much disputed Medicare "doc fix." But this time around would be potentially much worse. Billions of dollars worth of bills would go unpaid. There would be chaos, intense media attention, political hysteria, and probably lawsuits. The Treasury is going to have to roll over $500 billion in debt instruments during the period from February 15 to March 15. It will "have to pay higher interest rates to attract new buyers," and it is perhaps even "possible, if unlikely, that not enough bidders would appear."
Rating agency downgrades could also be more extensive and costly this time around. And the market risks beyond the X Date could ripple through to the equity markets (with effects on millions of people's pensions), as well as to our economy and the global financial system. "No guarantee of the outcome; risks are risks."