Expert Commentary

The short- and long-term effects of a U.S. government shutdown

Studies on the effects of the 2013 U.S. government shutdown, including changes in short-term spending habits and crime rates.

us capitol

An estimated 800,000 government employees have been affected by the partial government shutdown in the United States, which started on December 22, 2018. Roughly 380,000 federal workers have been furloughed and 420,000 are working without pay. The economy has taken a hit, and industries like commercial aviation are losing revenue. On January 15, Ed Bastian, Delta’s CEO, announced that the airline had lost $25 million in revenue on account of the shutdown. Grant reviews have been suspended by the National Science Foundation; national museums and cultural institutions have closed; and food banks have opened up for federal workers. Staff at national parks have been furloughed, and all fifty states have federal employees affected by the shutdown. (About one-fifth of federal workers are located in the Washington metro area; the rest are dispersed across the country.)

In short, the effects are far-ranging and palpable. And there’s much to learn about the potential short- and long-term effects of this shutdown — the longest in U.S. history — by looking at research on past shutdowns.

The 2013 U.S. government shutdown, which extended from October 1 to October 17 of that year, resulted in an estimated $24 billion in lost economic output. According to a Congressional Research Service report, 850,000 workers were furloughed at the shutdown’s start, and some 1.3 million essential employees worked without known payment dates.

Below we’ve summarized several peer-reviewed studies on the effects of the 2013 shutdown, which look at topics including short-term spending habits, changes in crime rates, fluctuating stock prices, long-term effects on employee morale and tourism to communities neighboring national parks.

“How Individuals Respond to a Liquidity Shock: Evidence from the 2013 Government Shutdown”
Gelman, Michael; et al. Journal of Public Economics, July 2018.

When government workers don’t get paid, how do their spending and saving habits change? This study looks at effects of the 2013 U.S. federal government shutdown, analyzing financial account records from 6,792 federal workers—including both those affected by the shutdown and not—as well as data from 91,692 non-federal workers.

Key findings:

  • A majority of government workers did not have a financial cushion sufficient to cover two weeks’ expenses.
  • For affected workers, “Weekly spending declined by roughly half the reduction in paycheck income and then recovered roughly equally over the two pay periods following the end of the shutdown.” The authors estimate for each dollar of liquidity lost, individuals spent about $0.58 less. They add that most people affected “reversed this drop in spending” after receiving their back-pay.
  • The researchers found that affected workers “smooth[ed] consumption using low-cost methods to shift the timing of payments for committed forms of expenditure. In particular, many affected workers delayed mortgage payments; and many shifted credit card balance payments.”
  • Affected workers did not increase their spending on credit cards. “Average debt only increased due to delays in debt payments.”
  • “While the data show that many affected workers were able to use unconventional means to smooth consumption, for some with low liquid assets these methods were either inadequate or unavailable. This group, who was carrying some credit card debt already, emerged from the shutdown with still more debt owing to failure to make payments rather than new borrowing.”

“What Happens When Employees Are Furloughed? A Resource Loss Perspective”
Baranik, Lisa E.; et al. Journal of Career Development, March 2018.

After the 2013 federal government shutdown, researchers gathered survey data from 212 furloughed employees to understand the personal consequences they experienced, in areas including life satisfaction and work-family conflict.

Key findings:

  • The furloughed employees experienced “decreased life satisfaction and increased work–family conflict and physical, cognitive, and emotional burnout 5 weeks after the shutdown ended.”
  • “Our findings show that furloughs can and do negatively affect employees and that these effects last long after the furlough has ended.”

“Politics, Bureaucracy, and Employee Retention: Toward an Integrated Framework of Turnover Intent”
Ali, Susannah Bruns. Administration & Society, February 2018.

This study draws on interviews with 35 federal employees representing 21 different agencies and three other political roles to offer qualitative insights on the effects of different factors on employee retention. One of the identified issues was “Shocks to the environment—effects of system breakdowns, furloughs, and busted budgets.”

Key findings:

  • “Interviewees indicated that short-term political plays have long-term effects on employee morale and career choices.”
  • For example, reflecting on the federal government shutdown in October ’13, interviewees expressed feelings of low morale and predicted turnover would increase. One interviewee said, “During the shutdown we were treated like a bunch of waste.”
  • Another said, “A lot of the people who hung on through the recession, and are feeling more secure, once we got the shutdown, they decided that they had had it. They were done. You can expect to see a rush to the door of the old civil service people. And then on the other side, some of the younger people are saying, ‘Whoa, what is this? Is this what I have to look forward to?’ That will lead some of them to start looking.”
  • Interviewees also expressed exasperation and a lack of trust in the federal budget process: “The furloughs and the shutdown were very damaging. We all just think it is going to happen again,” one subject said. Not having a budget, “That kind of thing wears on people, especially when you are in a leadership position, and you are responsible for resources and getting work done.”
  • Another theme in the interviews was the tension caused by selecting “essential” and “non-essential” employees. “Essential” employees are required to keep working without pay, whereas “non-essential” employees are furloughed. On the one hand, employees considered non-essential “were really offended thinking that others thought what they did was not important,” per one worker. On the other, some essential employees felt that “those who were nonessential got a paid 2.5-week vacation, while they held down the fort.” (Furloughed workers received back-pay after the 2013 government shutdown.)

“Federally Mandated Furloughs: The Effect on Organizational Commitment and Organizational Citizenship Behavior”
Shepherd, Robert. Defense Acquisition Research Journal, January 2017.

This study analyzes survey data collected from 42 furloughed federal employees during the 2013 government shutdown and 42 nonfurloughed active duty military personnel. Participants were asked questions that got at their commitment to their respective organizations (termed “organizational commitment” [OC]) and their organizational citizenship behavior (OCB), defined as “an employee’s actions contributing to and supporting a larger social and psychological atmosphere within the organization.”

Key findings:

  • Nonfurloughed employees had significantly higher organizational commitment (OC) scores than furloughed employees. In responses to all questions but one – “I would accept almost any type of job assignment in order to keep working for this organization” – furloughed employees ranked lower in terms of organizational commitment.
  • The biggest differences in responses between furloughed and nonfurloughed employees were to the following questions: “I talk up the organization to my friends as a great organization to work for” and “I find that my values and the organization’s values are very similar.” Furloughed employees scored these items significantly lower than nonfurloughed employees.
  • Nonfurloughed employees ranked higher than furloughed employees on all but three of 16 items related to OCB. “These items were ‘punctuality,’ ‘attendance at work is above the norm,’ and ‘make innovative suggestions to improve department.’”

“Income Changes and Consumption: Evidence from the 2013 Federal Government Shutdown”
Baker, Scott R.; Yannelis, Constantine. Review of Economic Dynamics, January 2017.

This study looks at data collected through an online personal finance website to analyze the effects of the 2013 government shutdown on spending. The sample included 91,650 unaffected government worker households and 61,160 affected government worker households.

Key findings:

  • “For affected workers, there is a small drop in the second period week of the shutdown, and a larger drop in the third week of the shutdown. This pattern is consistent with credit constraints and federal government workers exhausting their savings. The largest drop is seen in the third week of the shutdown, after the end of the shutdown was announced. This pattern is not consistent with alternative explanations such as the drop in spending being driven by revised beliefs about permanent income. Following the shutdown, there is a rebound in spending.”
  • The authors’ analysis finds that “the shutdown was associated with a 23.4 percent weekly drop in income for affected federal government workers relative to unaffected federal and state government workers.”
  • Affected federal government workers decreased their spending by approximately 10.7 percent compared to unaffected federal and state government workers.
  • Analyzing household spending patterns, the authors found that much of the decline for affected workers could be explained by “changes in leisure time and work-related expenditures.” For example, shopping, restaurant spending, home services spending, and public transit spending all increased after the shutdown.

“Effects of the October 2013 U.S. Federal Government Shutdown on National Park Gateway Communities: The Case of Acadia National Park and Bar Harbor, Maine”
Gabe, Todd. Applied Economics Letters, March 2016.

During the 2013 federal government shutdown, all 401 national parks were forced to close. This had the anticipated consequence of reducing visits to the parks in the near term. But what of the communities that neighbor these parks? This study analyzes the effects of the shutdown on Acadia National Park and the neighboring community of Bar Harbor, Maine.

Key findings:

  • In October 2013, Acadia had 94,904 visitors; down from the October average of 287,508 from 2010 to 2012 – a 76 percent reduction in visitation.
  • Tourism-related sales in Bar Harbor declined by 13 percent.
  • Comparing the decline in Acadia visits with tourism-related sales in Bar Harbor, the author concludes that “17.1 percent of the people who planned to visit Acadia during the shutdown cancelled trips to Bar Harbor.” The author writes, “Although results suggest that most of the potential Acadia visitors impacted by its closure still travelled to the Bar Harbor area, our estimate of about 17 percent of these visitors modifying their travel plans is a reasonably large response to a short-term event such as the government shutdown.” The author adds that if the shutdown had lasted longer, more tourists might have cancelled their trips.

“’Ain’t No Rest for the Wicked’: Population, Crime, and the 2013 Government Shutdown”
Gil, Ricard; Macis, Mario. Institute for the Study of Labor working paper, February 2015.

How do crime rates change in a city that’s been hollowed out by a government shutdown? This study, published by a private, independent economic research institute based out of Germany, compares crime rates in Washington, DC with those in Baltimore, Maryland to understand potential impacts of the 2013 federal government shutdown. The study also tests the approach by conducting the same analysis with crime data from DC exactly one year earlier. Types of crime studied in the analysis include burglary, robbery, theft and homicide.

Key findings:

  • The shutdown was linked to a 3 percent reduction in daily crime in DC.
  • While overall crime went down, crime rates increased in the evening (5 pm – 2 am). “The net effect was the result of a 9 percent decline of crime during working hours (9AM-5PM), a 5.6 percent decline during the early morning hours (2AM-9AM) and a 5 percent increase in crime during the evening and night hours (5PM-2AM). We also found that the decline in crime was concentrated during workdays (4 percent reduction), while crime overall increased during weekends (1.9 percent increase).”
  • The authors conclude that this is indicative of “reallocation of criminal activities” from days and times when the population of D.C. declined as a result of the shutdown.
  • The analysis of data from October 1 to October 16, 2012, which found no significant changes for crime rates, further corroborates the findings.

“Government Shutdown: A Test of Market Efficiency”
Woodward, Jessica H.; Bacon, Frank W. Journal of Business and Behavioral Sciences, fall 2015.

How does the stock market respond to government shutdowns? Not well, according to this research, which looks at the impact of the government shutdowns of 1995-1996 and 2013 on the stock price returns of a sample of 50 firms for each.

Key findings:

  • “Results demonstrate that the announcements of the 1995 and 2013 government shutdown had a significant negative impact on the firm’s share price up to 30 days prior to announcement day 0 with a continuous negative trend up to 30 following day 0.”
  • “The results for both samples suggest significant trading activity up to 30 days prior to the announcement of the government shutdowns possibly suggesting that investors may have anticipated the events.”

“Transportation Impacts of the 2013 Federal Government Shutdown in Northern Virginia”
Asare, Sampson; et al. Journal of Transportation Engineering, July 2015.

This paper looks at northern Virginia roadway volumes, high occupancy vehicle volumes, and transit ridership during the 2013 federal government shutdown. The authors compared data from a month before the shutdown to the two-week shutdown period and the two weeks after the shutdown.

Key findings:

  • All modes of transit saw reductions in use during the shutdown.
  • Transit impacts were larger in the second week of the shutdown compared with the first. Peak reductions ranged from 5.5 to 6 percent on arterials and freeways. Weekend reductions ranged from 7.3 to 8.5 percent, which the authors suggest might be linked to the closure of attractions like museums and monuments in the capitol.
  • High occupancy vehicle volumes and local bus service saw the largest reductions in use.
  • Revenues for the Washington Metropolitan Area Transit Authority, which provides subway and bus service for Washington, DC and its environs, declined by 3.73 percent during the first week of the shutdown and 10.91 percent during the second week of the shutdown.
  • Transit use largely returned to normal volumes after the shutdown.

About The Author