Debt Limit Impasse Could Drag On For Months
A showdown over the federal debt limit looms in the first half of 2011. Some Republicans in the Senate are threatening to block a debt limit increase if President Obama does not agree to sharp reductions in spending. If we actually hit the debt limit, Talking Points Memo warns that "all hell will break loose."
Even Americans for Tax Reform, whose president has previously expressed a desire to "drown government in a bathtub," is sending cautious notes about the debt limit, warning that "[n]o one regardless of who is in the majority wants to see the US default on its debt."
The consensus view seems to be that a debt limit impasse will lead to an acute crisis: at least a government shutdown, possibly a failure to make interest payments on government bonds, and an ensuing spike in bond yields and fall in stock prices. The urgency of the crisis would make it self-limiting, one side would quickly cave, and Congress would approve a debt limit increase.
I think another scenario is at least as plausible: failure to raise the debt limit could lead to a prolonged standoff, with the Obama Administration using financial strategies and accounting gimmicks to stay technically within the limit and continue operating the government. In doing so, the Administration would draw on plans created when the federal government hit the debt limit in 1995, and essentially copy tactics that are regularly used by illiquid state governments. Obama and the Congress could muddle through for months or, theoretically, years without a debt limit increase, and without an acute crisis in the bond markets.
So, while Dean Baker of the Center for Economic and Policy Research warns that a failure to raise the debt limit could lead to soaring Treasury bond yields and a Greek-like sovereign debt crisis, I think that probably would not happen. Even if Congress won't raise the debt limit, the United States government retains the economic capacity to pay its bills, and the markets understand that Congress will act to raise the debt limit if the other option is a financial collapse. By contrast, Greek bond yields are through the roof because Greece may be unable to pay its bondholders even if it wants to.
It's possible that legal uncertainty around the President's options for muddling through could lead to a sharp spike in interest rates despite the fundamental solvency of the federal government. If that happened, I believe the conventional wisdom would hold and a debt limit increase would quickly ensue, with no party wanting to be blamed for a debt crisis. This would cause bond yields to drop back to normal. But the certainty of such a response might mean that a spike in bond yields will be avoided altogether, especially if investors do not feel they have a better safe haven than Treasury bonds.
I'm inclined to bet against a spike in bond yields because we've been down this road before: the Wall Street Journal notes that the federal government hit the debt limit once in the last twenty years, and came close two other times. We hit the limit when the Gingrich-led Republican House refused, for a time, to raise the debt limit in 1995 (or rather, attached strings to a debt limit increase that President Clinton found unacceptable and vetoed); and we came close in 2002 and 2003, when split and Republican Congresses were reluctant to send a debt limit increase to President Bush.
But while the Journal has the history right, it wrongly says the government was "close to default" in these three cases. Not in the view of the bond markets; Treasury rates did not spike during the last three debt limit showdowns. Markets were confident that Congress would eventually raise the debt limit, and that in the meantime the President would be able to manage the government's liquidity without a default. The President and Congress cut deals to raise the debt limit fairly quickly, but not because the bond markets made them.
Let's review what happened last time we actually hit the debt limit, in 1995. The debt ceiling impasse led to a government shutdown, but that was because Congress and the President simultaneously could not reach agreement on a budget or appropriations bills; even if Congress had raised the debt limit, the Clinton Administration would have been sharply restricted in its ability to spend. In the runup to the government shutdown, the Washington Post reported on a long menu of options available for dealing with hitting the debt limit, many of which did not involve shutdown. Most of those options remain on the table for any upcoming standoff.
The most financially promising of those options would be to "disinvest" government trust funds that hold Treasury debt, most principally the Social Security Trust Fund. Essentially, the trust funds would redeem bonds they hold ahead of schedule, in exchange for a promise to be paid back later -- and such an IOU would not count against the debt limit. Because the trust fund balances exceed $2.5 trillion, this tactic could be used to run the government for several years without hitting the debt limit.
This tactic works because the debt limit applies to gross debt, including trust fund holdings, which are debts the federal government owes to itself. The government's true indebtedness is better reflected by net debt, also called Debt Held by the Public. By reducing the amount of debt that the government owes to itself, the federal government can grow the net debt while staying within the gross debt cap.
The Reagan Administration used this tactic during a budget standoff in 1985; while the AARP sued to block the raid on the Social Security Trust Fund, their suit was dismissed. However, Reagan used the proceeds of the raid to pay current Social Security benefits, and it may not be legal to raid the funds to pay for general operations of government. A raid would also be politically fraught; while the Social Security Trust Fund is an accounting fiction, most Social Security beneficiaries don't see it that way.
The 1995 Post article also provides a list of similar but smaller cash management options: borrowing from public employee retirement funds (a tactic that President Bush used when we nearly hit the debt limit in 2002 and 2003); recalling federal funds on deposit with commercial banks; borrowing from the Exchange Stabilization Fund; taking out a loan from the IMF (!); selling the government's gold reserves. However, these options would not buy as much breathing space as raiding the trust funds.
One other option is present now that was unavailable in 1995: some sort of manipulation of the Treasury debt that the Federal Reserve has purchased in the last two years as part of quantitative easing efforts. The Fed could forgive interest payments on this debt; since the Fed ultimately gives its returns on assets back to the federal government, this would not actually cost money, though it also wouldn't do that much to ease the government's cash flow crunch.
More radically, the Fed could forgive principal on the bonds it holds, which is to say it could monetize the debt. This would lead to inflation, which could be a feature or a bug, depending on the quantity. But such a move would also likely enrage members of Congress and add fuel to some conservatives' desires to rewrite the Federal Reserve Act. As such, I suspect the Fed would not endanger its independence by taking this step unless all other non-default options had been exhausted.
The last option the Post discusses is the one most frequently used by states facing cash flow crunches, and one that I would expect to be a feature of any prolonged debt limit standoff: "delaying payments to government contractors or federal employees." Such a strategy works because the debt limit applies to explicit bond debts, but not implicit liabilities, such as bills due and unpaid. By delaying contractor payments, the government behaves like a financially stressed business trying to stretch out its trade payables.
In 2009, when California legislators failed to pass a new budget, Governor Schwarzenegger paid state employees with IOUs instead of cash. Illinois regularly manages its cash imbalances by delaying payments to Medicaid providers. Periodically, when providers threaten that they will stop providing services unless the government ponies up, the state issues bonds to pay off the backlog. The federal government could use a similar strategy in Medicare to buy time while waiting for a higher debt limit. It could also delay other payments, such as aid to state governments.
Theoretically, the government could even avoid breaching the debt limit by delaying Social Security payments. Of course, this would be massively unpopular, though the Clinton Administration threatened to do it -- and the Obama Administration might be willing, if it thought it could saddle Republicans in Congress with the blame for missing checks.
These are bad fiscal practices, but they are better than bond defaults. States that use such tactics regularly, including Illinois and California, have retained access to the bond markets throughout. They do pay among the highest bond yields faced by state bond issuers, but it is not clear how much of that is due to liquidity-restricting budget impasses rather than fundamental questions about solvency. It seems to be possible to behave quite badly in liquidity management without spooking the bond markets very much.
Gimmicks like this can't work forever, and they probably can't work nearly as long as a raid on the Social Security Trust Fund. They would also have significant negative consequences: withholding grants in aid to states would worsen state budget crises, for example, and delayed payments to Medicare providers could make some medical practices insolvent.
But states have shown that while these gimmicks can't be used forever, they can go on for an awfully long time. Illinois's backlog of unpaid bills, mostly owed to Medicaid providers, currently stands at over $6 billion, or about half the state's total annual spending on Medicaid. Yet, Illinois's Medicaid program continues to operate and its beneficiaries continue to see doctors.
A delay in state aid payments could also be more viable than it sounds, if it simply leads states to issue more revenue anticipation bonds -- essentially using state borrowing to get around the federal debt limit. Such a move is also precedented by states' managing their books by delaying aid payments to municipalities.
One possible pitfall with a payment-prioritization strategy is that the President's legal authority to pick and choose bills to pay is unclear. In 1995, Republicans in Congress offered to pass a bill giving the President explicit authority to do this. Absent such a bill, the President's default-avoidance strategies could end up in court -- but Congress might again be willing to offer the President such powers, as failing to do so would essentially be an explicit demand for a default on the public debt.
In light of the options I've sketched out, let's think about how a debt limit impasse might actually play out. A failure to raise the debt limit need not lead to "all hell breaking loose," any more than we would say that all hell has broken loose in Sacramento or Springfield. Instead, a failure to raise the debt limit could drive the Administration to use a series of irresponsible and somewhat painful cash-management actions that would nonetheless keep the government's head above water, service the debt, and even largely avoid a shutdown of government, for months.
The downside of a debt limit impasse being a non-catastrophe is that Washington lawmakers are more likely to allow it to happen. Even though my scenario does not include a near-term default on the public debt, it could increase the long-term risk of a sovereign debt crisis by making the federal government appear to be a less responsible and creditworthy debt issuer -- essentially, it could be a start of the Sacramentoization of federal finances. This would be a very bad thing, though we would not feel its effects immediately.
So, for the sake of our government's long-term financial health, I hope that Republicans and Democrats are able to come to terms and raise the debt ceiling when necessary. But lawmakers are generally less responsible about responding to long-term problems than short-term ones. For that reason, I worry that Washington politicians may subject us to a months-long, slow-motion liquidity crisis in 2011 -- and that the comedy of errors that is California's fiscal management may go national.