As we explained first in January 2011, there is nothing sinister about this. Any time the Treasury hits its physical debt cap, it activates its available "emergency measures" which include such money releasing options as disinvesting the Civil Service Fund, Suspending reinvestment in the G-Fund, Selling securities from the Exchange Stabilization Fund, and others, which cumulatively free up around $300-$350 billion. In essence the "emergency measures" act like a revolving credit facility that is slowly but surely being drawn down. Add to that sporadic cash creation over the past few months from cash inflows from the GSEs and one can see why the US has been able to be in breach of the debt ceiling for as long as it has. And why it still has just under two months of capacity.
The chart below shows how much emergency capacity the US has currently when adding all the emergency benefits, and how much it will have over the next three months. Of note is the period between October 15 and November 1, when tifrst the "revolver" cash balance dips below $50 billion, and then hits $0 by November 1. Either way, there better be a debt deal by mid-October, or late October at the very latest, or the summer of 2011 debt ceiling fiasco will seem like a walk in the park in comparison to what's coming.
Q: What did Treasury announce earlier this week?
Treasury Secretary Jack Lew informed Congress on August 26 that the Treasury would exhaust its borrowing capacity by the "middle of October." Specifically, this is when the Treasury has determined that it will have used the last of the "extraordinary measures" it has been employing to temporarily lower debt held in internal trust funds in order to make room under the debt limit for marketable Treasury issuance. Once those accounting strategies are no longer available, the Treasury will be left to rely on incoming tax receipts and the cash balance on hand to finance any disbursements.
The Treasury projects its cash balance to be roughly $50bn in mid-October when it exhausts its borrowing authority. This is a greater level of detail than the Treasury has provided ahead of previous debt limit deadlines. Prior announcements never stated the precise cash level the Treasury expected to have around the deadline, and thus left some uncertainty about whether payments scheduled to be made in the days following the deadline were truly at risk. For example, the Treasury announced in 2011 that it expected to exhaust its borrowing authority by August 2 of that year. Deficits turned out to be slightly lower than anticipated in the weeks leading up to the deadline, leaving the Treasury with $54bn in cash on hand at the deadline, which could have allowed the Treasury to operate without additional borrowing for at least another week.
Q: How does that compare with our forecast?
We had previously projected that Congress would need to raise the debt limit before November 1, give or take a week or two. However, although the deadline we projected differs from the Treasury's "middle of October", it is the definition of the deadline that differs rather than our view of projected Treasury cash flows. Our latest projection implies that the Treasury's total financing capacity--i.e., the headroom under the debt limit plus the cash balance--will drop below $50bn in mid-October, similar to the Treasury's projection (Exhibit 1). Based on the pattern of daily receipts and outlays that we project (we base these on prior years' patterns scaled to current levels and adjusted for calendar differences), we expect that the roughly $50bn cash balance that the Treasury will have in mid-October will gradually shrink over the following two weeks. The Treasury would probably be able scheduled payments in late October with the cash we expect it to still have on hand, but it would be unable to make all of the payments scheduled for November 1.
Q: So what is the real deadline?
To avoid disruptions to the Treasury market, Congress will probably need to raise the deadline by mid-October, as the Treasury states. Since the Treasury plans its securities issuance with the expectation of a sizeable cash cushion--the cash balance has averaged around $60bn over the last couple of years--it should be expected that the debt limit, net of extraordinary measures, would be reached before the point that the Treasury is unable to pay its obligations as they come due. The upshot is that Treasury's regularly scheduled auctions could be scaled back or delayed if Congress doesn't raise the debt limit by Treasury's stated deadline, even though scheduled payments might not be immediately at risk.
The Treasury is scheduled to hold auctions of 3- and 10-year notes and 30-year bonds the week of October 6 that are scheduled to settle October 15. The Treasury's projection that it will exhaust its "extraordinary measures" by mid-October seems reasonable if one assumes that Treasury auctions continue on schedule in amounts similar to what is implied by the Treasury's most recent issuance projections. Assuming those projections are accurate, the Treasury might not be able to conduct those auctions as planned without a high degree of confidence that the limit would be raised by the time the securities settle on October 15.
This would be highly unusual but not unprecedented. A Government Accountability Office (GAO) report lists 17 Treasury auctions between 1995 and 2006 that for which the announcement date was delayed; in 11 of those cases the auction itself was delayed. Most of these involved short delays to bill auctions, but during the 1995 debt limit debate the Treasury postponed 3-year and 10-year auctions for roughly one week.
Q: How does the Treasury's announcement affect the upcoming fiscal debate?
The announced deadline probably shifts attention further toward the debt limit and away from the debate over extending government spending authority. Current spending authority expires at the end of the fiscal year on September 30. While that issue could still be dealt with separately from the debt limit, a mid-October deadline increases the likelihood that they will be rolled up into one debate given their proximity on the calendar (perhaps involving a short extension of spending authority to more closely line up the deadlines). The practical consequences of this are not that great, however. Republican leaders have not appeared enthusiastic about strategies to block the renewal of spending authority unless changes to the Affordable Care Act (also known as "Obamacare") are made, and had already appeared to be more focused on negotiating around the debt limit even before the Treasury's announcement.
The detail the Treasury has provided regarding its cash balance may also lead some lawmakers to believe they have more time to raise the limit than they really do. As noted earlier, the Treasury's mid-October deadline is analogous to the August 2 deadline from 2011. The Treasury still had a significant cash balance it could use to make scheduled payments, but due to normal scheduled Treasury issuance it had exhausted its capacity to borrow under the limit. In the 2011 episode, lawmakers had little doubt that they needed to raise the debt limit by the deadline. This was, among other things, due to the perceived risk that the Treasury could miss an important scheduled payment, which seemed quite possible without knowledge of how much cash the Treasury expected to have on hand. If lawmakers feel confident that payments scheduled for the days following the mid-October deadline are likely to be made on schedule, they may feel less pressure to enact an increase by that point and could conceivably wait until later in the month. (Complicating matters further, Congress is scheduled to be on recess October 14-19.) Given this risk, it seems likely to us that the Treasury will make clear to Congress in upcoming communications that while it may have some remaining cash on hand to pay obligations after the mid-October deadline, waiting until late October would nevertheless be disruptive and should be avoided.