Lawmakers face yet another deadline over the federal budget on Thursday, raising the possibility of a second consecutive government shutdown. Amid this latest round of budget negotiations, the debt limit “X Date” is quietly approaching (BPC’s latest projection is March 2018), at which point the federal government would no longer be able to meet its financial obligations in full and on time. Given the proximity of those two events, if an agreement on budget levels and other key priorities cannot be reached, it will be important to know how a government shutdown could interact with the debt limit.
A prolonged shutdown at this time during tax season could have an impact on the “X Date,” primarily because it could delay the processing of 2018 tax refunds. However, our analysis of recent government shutdowns shows that to have a large enough impact to materially move the “X Date,” a government shutdown would have to last at least a week, possibly longer. According to our analysis, other categories of spending do not fall significantly enough to have a large “X Date” impact.
Unlike other recent shutdowns, the impending one would land squarely in the middle of tax season. Given IRS guidance that, in the case of a shutdown, employees in charge of issuing tax refunds would be furloughed, any extended closure would likely slow down the distribution of refunds. Tax return filing started as early as Jan. 29, and any refunds owed are typically delivered by the IRS within 21 days. If disbursement of refunds takes longer than usual, it could push the “X Date” later, because the government would be holding those funds longer.
In the case of a shutdown, employees in charge of issuing tax refunds would be furloughed, any extended closure would likely slow down the distribution of refunds.
Last year, individual tax refunds totaled about $180 billion during February and March, so the pool in question is large enough to have an impact if a portion of it were delayed by several weeks. The longer a government shutdown continues the more refunds would be delayed, potentially pushing back the “X Date.”
Refunds are the key question mark when it comes to a potential shutdown’s impact on the “X Date,” because our analysis of other spending showed only small effects. Using the Treasury Department’s daily statements, we examined key categories of spending, comparing their levels during the shutdown to the equivalent days in the years that came immediately before and after. The most recent shutdown of significant length, which took place in 2013 from October 1 to 16, began on the first day of the fiscal year and lasted for just over two weeks. Overall, most categories of spending experienced little-to-no decline compared to the same period in 2012 and 2014.
10/1/12 - 10/16/12
10/1/13 - 10/16/13
9/30/14 - 10/15/14
Source: Daily Treasury Statements; BPC Calculations
Specific categories of spending that one might assume would be affected by a shutdown only show relatively small changes. Funding for unemployment insurance is mandatory (i.e., not annually appropriated), so this spending should have been unaffected by the shutdown. Indeed, the drop in spending on unemployment insurance over the three years roughly mirrored the downward trend in the number of new unemployment insurance claims, which fell from 330,000 during the first week of October 2012 to 258,000 in October 2014. Even categories of spending that are primarily discretionary, such as highways and energy, did not see a noticeable decrease, likely because certain spending is exempted by law from the shutdown or was already appropriated in prior years.
The biggest exception is that spending on federal salaries did decrease from $8.7 billion in 2012 to $6.2 billion during the shutdown year. While some of the difference could be explained by natural variation (or sequestration), furloughs from the shutdown almost certainly contributed. Overall, there is a strong case that spending fell as a result of the 2013 shutdown, but the magnitude was relatively small.
These effects are similar to what we found when analyzing the three-week shutdown from December 1995 to January 1996. Although it lasted longer, spending levels during this period saw even less fluctuation than in 2013 because seven appropriations bills were passed before the shutdown began.
12/19/94 - 1/6/95
Source: Daily Treasury Statements, Courtesy of Haver Analytics; BPC Calculations
The bottom line: if a shutdown materializes, it would need to last for an extended period into tax refund season to have a material impact on the “X Date.” Changes in spending not related to tax returns would only have a modest impact on Treasury’s cash flows. The spending effects of a short-term government shutdown is likely already captured by fluctuation within BPC’s “X Date” range, and does not warrant an entirely new projection. BPC plans to issue an updated projection soon, taking into account the effects of the new tax law on the “X Date.”