So, the government has shut down. Now what?
The government in fact did shut down, which was a bit of a surprise to many who expected a last minute deal. However, contention is so pervasive in both houses that they just couldn’t get there. While essential services, such as the military, law enforcement, air traffic control and asteroid watch are still happening, other ‘non-essential’ government programs remain indefinitely closed, which is taking a small bite out of the economy.
What is the probable outcome/timetable? That is, of course, dependent on the parties involved and is anyone’s guess. But the work of several well-regarded economists and political strategists points to a combined resolution during the week prior to the Oct. 17 debt limit deadline, largely because there is little pressure to do so beforehand. Or, another scenario involves taking care of the debt limit as a separate issue while keeping the government shutdown in effect. That’s certainly not ideal, but a preferable option to not addressing anything.
The debt limit increase legislation, which allows the government to continue to fund itself, and more importantly, pay interest on existing debt and avoid default (either an outright default or one caused by a delay in payment). Neither is acceptable from a market standpoint, as payment of interest on debt gets to the core credibility of any borrowing entity, and not fulfilling this obligation would be extremely disruptive—from the primary event itself as well as secondary repercussions. These could include further rating agency downgrades, which affect the ability of certain institutions to continue to hold Treasury debt as ‘risk-free,’ which, if threatened, could create other market selling pressures. We’d rather not go there. Then again, we almost went there in 2011 and Treasuries rallied while stocks tanked—an episode that still baffles economists and bond portfolio managers we speak to regularly. By contrast, the economy in 2013 is stronger than it was in 2011 and we don’t have the same level of overhang around European sovereign debt issues, which added to anxiety that summer.
What about the underlying economic impact? Initially, it was thought a one-week shutdown could reduce this quarter’s GDP by 0.20-0.25% (including roughly $400 million in federal compensation per day), based on estimates, with a two-week event perhaps a bit less than double that amount. There is a linear reduction in GDP the longer this goes on. A quarter or half a percent isn’t a huge number generally, but at current low rates of growth (2-3%), it could start to become significant. However, now that the defense department is being largely called back in from furlough, the net effect may fall to 0.15% a week. The final erosion to growth may also change a bit based on any federal payrolls made retroactive to the beginning of the furlough.
The conundrum stems from this: while the government is technically saving money in the budget by not paying employees to work, this naturally translates into these employees consuming less in the economy (payrolls usually being spent, aren’t). From an accounting standpoint, lower payrolls mean the government itself is ‘consuming’ less. Retroactive payroll legislation could certainly help reverse some of this.
What’s the worst that could happen? Congress doesn’t agree to raise the debt limit, which might result in suspended Treasury auctions and ‘payment prioritizations’ to ensure Treasury interest is paid before other obligations. There is a quirky precedent for this—in 1957, when the debt limit wasn’t raised immediately, the Treasury delayed payments to federal contractors. In early 1996, amidst the last big government shutdown, a separate bill to ensure payment of Social Security was put in place before the debt limit was officially increased after threats of payments being suspended. How practical a ‘payment prioritization’ plan might be considering the complexity of the federal payment system is another question entirely, but again, we hope we don’t have to go there. It’s important to note that a default of some sort, even a technical one, is not in the base case of most economists or strategists—it is considered far too damaging.
Markets have not reacted well to the shutdown so far, but this isn’t unexpected or uncommon. How does the market usually react during these closures? Not half bad, when it’s all said and done. In looking at the last 17 shutdowns that have happened since 1976, returns become progressively negative during the shutdown period (assuming it lasts at least several days), but things get back to normal as the situation resolves itself and the giddiness takes over.
In the last example during the 1995-1996 period (a total of 28 days from 11/14/95 to 11/19/95 and again 12/16/95 to 1/6/96), the S&P fared worst during the first few days of the crisis, but righted itself by the time the inevitable resolution came around. Again, fears of the unknown were eventually overcome by the less impactful reality—a trend we see again and again with crisis events.
Bottom line: This will be solved at some point. Based on historical precedent, selling into the chaos may be a bad idea as sentiment usually improves. Eventually.