The federal government experienced a funding gap beginning on October 1, 2013, which ended when the Continuing Appropriations Act, 2014 (P.L. 113-46) was signed into law on October 17, 2013. This funding gap resulted in a “government shutdown” and the furlough of federal employees who were not excepted. The Continuing Appropriations Act, 2014 also temporarily suspended the statutory debt limit through February 7, 2014. This report discusses the effects of the FY2014 government shutdown on the economy.
The government shutdown had both direct and indirect effects on economic growth. It directly reduced gross domestic product (GDP) because government spending is a component of GDP. The Bureau of Economic Analysis estimates that real GDP growth was reduced by 0.3 percentage points because of the reduction in hours worked by federal employees; it has not estimated the overall effects of the shutdown. Overall, federal spending fell by 6.6% in the fourth quarter, but this is comparable to the rate in previous quarters. Assuming the funding levels enacted on October 17, 2013 were the same as the funding levels that would have been enacted on September 30, 2013 had a shutdown not occurred, some spending would have been delayed, but not permanently reduced. For example, furloughed federal employees were paid in full, but late. Since the shutdown occurred at the beginning of the quarter, much of the delayed spending may have occurred before the quarter is over. An example of a potential indirect effect is a reduction in private consumption or business investment because of a decline in consumer confidence, which surveys reported in October. GDP growth rose to 3.8% in the fourth quarter of 2013, so it does not appear that the indirect effects were significant. Some of the indirect effects may be attributable to the debt limit impasse, however, which occurred at the same time as the shutdown.
The Administration reported that federal worker furloughs over the course of the shutdown were equivalent to 6.6 million work days, peaking at 850,000 workers furloughed at the beginning of the shutdown. The number of private employees laid off or furloughed as a result of the shutdown is unknown. Initial unemployment claims surged during the shutdown. For technical reasons, the shutdown had no effect on the more frequently cited job growth figure, but the Bureau of Labor Statistics reported that 440,000 federal workers were unemployed on temporary layoff or not at work for “other” reasons in October 2013 (compared to 15,000 in October 2012).
A review of third-party estimates of the effects of the shutdown on the economy finds a predicted reduction in GDP growth of at least 0.1 percentage points for each week of the shutdown, with a cumulative effect of up to 0.6 percentage points in the fourth quarter of 2013. Most of the estimates are predictions by professional forecasters, many of which were made before the shutdown had ended. These estimates are subject to error and uncertainty. Overall, most forecasters predicted that the effects of the shutdown on the economy would be small relative to the overall economy because it ended after 2½ weeks, it only affected a subset of federal spending, and it generally delayed rather than eliminated federal spending. The economic effects of the shutdown were unevenly distributed across regions, industries, or individuals, however. Where detail was provided, most forecasters did not factor in any multiplier or indirect effects of the shutdown. In that sense, the estimates reviewed can be thought of as a lower bound on the overall effects on economic activity.
The Congressional Research Service does not plan to provide an independent estimate of the economic impact of the shutdown. This report does not provide background on or explanation of recent or historical shutdowns or funding gaps. For information about government shutdowns, see CRS Report R43250, CRS Resources on the FY2014 Funding Gap, Shutdown, and Status of Appropriations, by Justin Murray and CRS Report RL34680, Shutdown of the Federal Government: Causes, Processes, and Effects, coordinated by Clinton T. Brass.
The federal government experienced a funding gap beginning on October 1, 2013, which ended when the Continuing Appropriations Act, 2014 (P.L. 113-46) was signed into law on October 17, 2013. This funding gap resulted in a "government shutdown" and the furlough of federal employees who were not excepted.1 P.L. 113-46 provided for all federal employees to be retroactively paid as if they had been at work for the shutdown period.
The Continuing Appropriations Act, 2014 also temporarily suspended the statutory debt limit through February 7, 2014. Treasury had predicted that it would exhaust its borrowing capacity under "extraordinary measures" by October 17, 2013. Federal debt outstanding had reached its statutory limit of $16.699 trillion on May 19, 2013, and the Treasury had relied on these extraordinary measures to fund government operations since then. When observing economic and financial data during the shutdown period, it is difficult to disentangle the effects of the shutdown from those of the debt limit impasse because they occurred simultaneously.2
This report discusses the effects of the FY2014 government shutdown on economic output. Gross domestic product (GDP) measures the value of goods and services produced in the economy. The shutdown could have potentially affected all components of GDP—government spending directly, and private consumption, capital investment, and net exports indirectly. It also reviews third-party estimates of the effects of the shutdown on the economy. The Congressional Research Service (CRS) does not plan to provide an independent estimate of the economic impact of the shutdown.
A government shutdown directly affects GDP because government spending is a component of GDP. Economic output, as measured by GDP, is based on the equilibrium between total supply (production) and total demand (spending). The FY2014 shutdown reduced both—it reduced total supply because furloughed government workers could not contribute to the production of government output, and it reduced total demand because certain government purchases of private-sector goods and services could not be made. The reduction in demand was temporary and would be reversed once delayed purchases were made. The reduction in supply was mostly unrecoverable (since lost working hours cannot be made up), but was a one-off effect that ended when work resumed. The Bureau of Economic Analysis reported that it could only isolate the effects of the shutdown on hours worked by federal employees, which it estimates reduced real GDP growth for the fourth quarter by 0.3 percentage points.3
Assuming the funding levels enacted on October 17, 2013, were the same as the funding levels that would have been enacted on September 30, 2013, had a shutdown not occurred,4 government spending, and hence GDP, would have been directly reduced for at least the length of the shutdown (perhaps longer, as some agencies might face a spending backlog), and then boosted after the shutdown ended. For example, under P.L. 113-46, all affected federal employees were paid in full for the length of the shutdown, albeit later than usual. Since the shutdown occurred at the beginning of the fourth quarter, with 2½ months remaining in the quarter for agencies to catch up, the fall and subsequent rise in government spending may largely cancel out within the fourth quarter.
It is unknown how much of total federal spending was delayed by the shutdown, with discretionary spending mainly affected. As measured in the GDP accounts, federal output of goods and services shrunk 6.6% in the fourth quarter and reduced overall GDP growth by 0.4 percentage points. While this decline is large, it cannot be wholly attributed to the shutdown—in the four preceding quarters, federal output shrunk by an average of 6.6%. Moody's estimated that one-fifth of total federal spending was affected by the shutdown for its duration.5 Total federal spending was about 20% of GDP in FY2014, so one-fifth would be equivalent to around 4% of GDP. Thus, the effect on GDP was significant for the duration of the shutdown, but relatively small on an annual basis since the shutdown ended after 2½ weeks. Nevertheless, the shutdown reportedly created hardships for some of those directly affected by it. In particular, the economic impact of the shutdown was not distributed evenly across the country—the Washington, DC, metropolitan area and other localities dependent on federal funds (e.g., the tourism industry near national parks that were shut down) were disproportionately affected.6 And since GDP only measures the value of output, it does not capture any costs associated with greater spending by state governments, lost federal user fees (e.g., entrance fees at national parks), changes in the value of financial securities, or private inconvenience costs associated with the unavailability of government services during the shutdown.
A shutdown could also have indirect effects on GDP. First, lower government spending is believed to have "multiplier effects" that make a decline in GDP greater than the associated decline in government spending. For example, if federal employees reduced consumption in response to delayed pay or federal contractors reduced investment in response to delayed payments, then those categories of GDP could also fall. A recent working paper found evidence that this was the case in 2013.7 The magnitude of multiplier effects is uncertain—CBO estimates that $1 of reduced federal spending on goods and services could reduce GDP by between $0.40 and $1.90 over the next four quarters if the Federal Reserve adjusts monetary policy in response, for example.8 Multiplier effects are temporary by nature and, as with the direct effects on government spending, the limited duration of the shutdown may mean that multiplier effects mostly resulted in delayed (rather than permanently reduced) private spending, with negative effects being offset by lagged spending in the same or next quarter.
Second, a shutdown could create dislocations for some individuals or markets that had broader effects on the economy. Moody's identifies exports and imports (because of delays in obtaining federal permits), mortgages (because of delays in verifying federal records), and small business loans backed by the Small Business Administration as examples of privately economic activity temporarily disrupted by the shutdown.9 In addition, the delay in salary to workers and payments to contractors could cause an uptick in delinquencies that would affect the creditworthiness of those individuals and firms going forward. If a shutdown caused a decline in the value of certain financial assets, it could have a wealth effect that affected household and business spending. It is difficult to reliably disentangle the effects of the FY2014 shutdown on financial markets from those of the debt limit impasse. To the extent that the shutdown affected financial markets, some immediately identifiable effects did not seem large or long lasting enough overall to have a discernible impact on GDP. For example, the S&P 500 stock index fell by more than 2.5% during the shutdown, but recovered its previous value by the time the government was reopened.
Finally, if the FY2014 shutdown (and debt limit impasse) caused a decline in consumer, business, or investor confidence, it could have led consumers and businesses to postpone or cancel spending decisions, particularly large orders for consumer durables or capital investment. Various private groups produce confidence indices based on surveys. The University of Michigan's consumer confidence index reportedly fell to a 10-month low in October 2013.10 Since 2008, Gallup has produced a weekly economic confidence index. For the week ending October 13, 2013, it reached its lowest point since early December 2011.11 For both the University of Michigan's and Gallup's indices, the October 2013 remained above their levels in October 2008, at the acute point of the financial crisis, and August 2011, following the previous debt limit impasse and downgrade of U.S. Treasury securities. The Conference Board's consumer confidence index fell from 80.2 in September 2013 to 71.2 in October 2013, but remained higher than it was in January 2013, during the "fiscal cliff" episode.12
A shutdown (and debt limit impasse) could negatively affect confidence and financial markets because they create uncertainty about future policy as well as overall financial conditions, given the potential financial effects of a failure to raise the debt ceiling. Economists generally believe that uncertainty is not conducive to economic growth. An isolated, temporary event would not be thought to have any lasting effect on confidence or uncertainty, but arguably because there have been recurring episodes of uncertainty about resolving the debt limit and fiscal policy with no permanent resolution, individuals could view the FY2014 episode as one in a series that will continue in the future. Viewed as part of a persistent pattern, such events could lead to lasting effects on confidence and uncertainty. In Moody's view,
Even if lawmakers come to terms roughly as expected, political vitriol and repeated threats to shut government or not pay its bills have weighed heavily on sentiment and meaningfully harmed economic growth. Businesses are more reluctant to invest and hire, and entrepreneurs are less likely to attempt startups. Financial institutions are more circumspect about lending, and households are more cautious about spending. While many factors are at work here, Washington's heated budget battles are a significant contributor. Half the CEOs in the Business Roundtable's third-quarter outlook survey said Washington's battles have affected their hiring plans over the next six months. Shaky nerves stifle risk-taking and entrepreneurship, which is key to stronger growth.13
The Government Accountability Office (GAO) reported that the economic forecasters it interviewed believe that the indirect effects of the FY2014 shutdown on the economy were minimal and did not anticipate that the shutdown would have any long-term effects on the economy.14 The direct and indirect effects were not significant enough to knock the economy off its growth path. Had the economic recovery been very weak prior to the shutdown, the shutdown might push the economy back into recession. Alternatively, had the economy been on a very strong growth path, other spending might be sufficient to absorb the impact of the shutdown, maintaining a rapid pace of growth.15 As it turned out, GDP growth was 3.8% in the fourth quarter of 2013, which was higher than the average growth rate and higher than growth in the three quarters preceding it and the following quarter. One reason that the FY2014 shutdown did not cause a recession was because it was relatively short; Moody's predicted that a shutdown exceeding two months would have pushed the economy into recession.16
The FY2014 shutdown directly affected the labor force through the furlough of non-excepted federal employees.17 According to the Administration, 6.6 million federal work days were lost to furloughs caused by the shutdown, peaking at 850,000 (equal to 40% of the federal workforce) at the beginning of the shutdown and declining over the course of the shutdown.18 On October 7, 2013, the Department of Defense (DOD) recalled almost 400,000 furloughed employees to work.19 These figures compare with total employment of 144 million, making the 850,000 estimate equivalent to about 0.6% of total employment. In addition to federal workers, private firms with federal contracts furloughed or laid off an unknown number of workers as a result of the shutdown.20
Data on employment and unemployment are released by the Bureau of Labor Statistics (BLS). Labor force data are based on a survey conducted once a month. In October 2013, the survey covered the period including October 12, and thus occurred during the shutdown period. BLS produces unemployment estimates in the "household" survey (the Current Population Survey) and two different estimates of employment—one from the household survey and a more frequently cited one from the "establishment" survey (the Current Employment Statistics). According to BLS, federal workers were classified as unemployed on temporary layoff in the household survey if they were furloughed for the entire survey period (regardless of whether they were retroactively paid), but they will be considered employed in the establishment survey because they were paid retroactively.21 Thus, the shutdown had no effect on the more frequently cited job growth figure, but increased the unemployment rate.
BLS reported that 437,000 federal workers were unemployed, of which 223,000 were on temporary layoff in October 2013. Some of these workers were unemployed or on temporary layoffs due to factors unrelated to the shutdown, however. To get a sense of how many federal workers were unemployed or temporarily laid off for reasons unrelated to the shutdown, Table 1 compares the number unemployed or laid off in October 2013 with September 2013 and October 2012.22 In addition, BLS reported that 217,000 federal workers reported their status as "not at work" for reasons other than one of the nine standard reasons, such as vacation or illness. According to BLS, these may have been furloughed workers who were erroneously reported their status as not at work instead of as unemployed or temporarily laid off; it is not BLS practice to correct erroneously reported data. As noted above, the number of furloughed workers declined over the course of the shutdown; workers who had previously been furloughed but were called back to work by the time the BLS survey took place were classified as employed, with some classified as employed part-time for economic reasons.23
Table 1. Labor Force Status of Federal Workers Before and During the Shutdown
(thousands of workers; not seasonally adjusted)
|
Month |
Unemployed |
Of which, Temporary Layoff |
Not at Work |
|
October 2013 (Shutdown) |
437 |
223 |
217 |
|
September 2013 (Comparison) |
270 |
19 |
n/a |
|
October 2012 (Comparison) |
244 |
6 |
9 |
Source: BLS.
Notes: BLS has not adjusted data for seasonal variations between September and October 2013. Workers who report their status as "not at work" can choose one of nine standard reasons or "other reasons;" the latter is shown in the table.
Data for initial unemployment insurance (UI) claims covering part of the shutdown period are also available.24 Initial claims by federal employees rose from 1,391 in the week ending September 28, 2013, to 70,071 in the week ending October 5, 2013, before falling to 14,423 in the week ending October 19, 2013.25 This is comparable to the change in overall initial claims over this period.
Beyond federal furloughs, lower GDP growth would be expected to slow the pace of employment growth temporarily. To the extent that lower GDP growth occurred in the federal government sector, lower employment growth may also be concentrated there. Except where noted, most of the economic estimates summarized in the next section did not include numerical estimates of employment effects. Overall, the unemployment rate was 7.2% in October 2013—unchanged from the previous two months.
Various economic forecasters predicted how much the shutdown would reduce economic growth. Many predictions were made before the shutdown had ended and were based on some assumption about how long it would last. GAO did not identify any pending or completed studies after the shutdown ended that looked back at the economic effects of the shutdown.26 In most cases, a detailed description of how these estimates were produced was not provided; generally speaking, forecasters would use their macroeconomic models to project GDP under two scenarios, a base scenario and a shutdown scenario, and compare the difference. Economic models are subject to error and uncertainty, particularly in relation to some less directly observable effects of the shutdown, such as its effects on confidence and how changes in confidence translate to economic activity.
All of the forecasts reviewed found that a relatively short shutdown would not substantially change the economy's growth path. The fact that GDP growth was 3.8% in the fourth quarter of 2013—higher than average and stronger than the three preceding quarter and the following quarter—supports this view.
Where detail was provided, most forecasters limited their analysis to the direct effects on government spending, and did not include any multiplier effects or indirect effects, such as effects on private confidence. Thus, these estimates can be thought of as a lower bound compared with estimates with a broader scope.
For context, 1% of GDP is equivalent to about $170 billion. But since quarterly growth figures are reported on an annualized basis, a percentage point change in quarterly growth is equivalent to about $40 billion.
Author Contact Information
Acknowledgments
The author would like to thank Jared Nagel, information research specialist, for research assistance.
| 1. |
For information about government shutdowns, see CRS Report R43250, CRS Resources on the FY2014 Funding Gap, Shutdown, and Status of Appropriations, by [author name scrubbed] and CRS Report RL34680, Shutdown of the Federal Government: Causes, Processes, and Effects, coordinated by [author name scrubbed]. |
| 2. |
The economic and financial effects of the debt limit impasse are discussed in CRS Report R41633, Reaching the Debt Limit: Background and Potential Effects on Government Operations, by [author name scrubbed] et al. |
| 3. |
Bureau of Economic Analysis, Technical Note, January 30, 2014, http://www.bea.gov/newsreleases/national/gdp/2014/pdf/tech4q13_adv.pdf. Government spending's contribution to GDP is calculated based on inputs, unlike private-sector output, which is based on value added. Nominal government spending is based partly on federal employee pay, so that the decision to pay furloughed workers results in the shutdown having no effect on nominal GDP. In contrast, real (i.e., inflation-adjusted) government spending is based partly on federal employee hours worked, so real GDP is reduced during the shutdown. Thus, the loss in real government spending due to reduced hours worked was mostly not be made up after the shutdown ends. This also means that the disparate measurement of furloughed workers in nominal and real GDP was reconciled by a temporary rise in the price index for government spending. For an explanation, see Macroeconomic Advisers, "Showdown over a Shutdown," September 25, 2013, p. 2. |
| 4. |
The Continuing Appropriations Act (P.L. 113-46) did not reduce levels of budget authority to account for the period when the government was shut down. The shutdown would lead to a permanently lower level of government spending only if the shutdown caused Congress to enact lower spending levels through January 15 than they otherwise would have (or than forecasters had previously projected). |
| 5. |
Mark Zandi, "U.S. Macro Outlook: Back to the Brink," Moody's Analytics, October 7, 2013. |
| 6. |
Estimates on which states were most affected by the shutdown can be found in Mark Zandi, "A Budget Battle Postmortem," Moody's Analytics, Dismal Scientist, October 21, 2013. Stephen Fuller, Director of George Mason University's Center for Regional Analysis, was quoted as estimating that the Washington metropolitan economy "could lose an estimated $200 million a day" from the shutdown. See Brigid Schulte and Justin Jouvenal, "Washington Area Could Lose $200 Million A Day if Shutdown Occurs," Washington Post, September 29, 2013. |
| 7. |
Michael Gelman et al, How Individuals Smooth Spending: Evidence from the 2013 Government Shutdown Using Account Data, National Bureau of Economic Research, Working Paper no. 21025, March 2015. The authors found that "Those with the least liquidity struggled most to smooth spending and were left holding more debt months after the shutdown." |
| 8. |
See Felix Reichling and Charles Whalen, Assessing the Short-Term Effects on Output of Changes in Federal Fiscal Policies, Congressional Budget Office, Working Paper 2012-08, May 2012, p. 3. If the Federal Reserve does not adjust monetary policy in response, CBO estimates that $1 in spending will have a multiplier effect of $0.50 to $2.50. The current scenario, where interest rates are already at the "lower bound," might be considered a situation in which the Fed could not effectively adjust policy to fiscal tightening. |
| 9. |
Mark Zandi, "A Budget Battle Postmortem," Moody's Analytics, Dismal Scientist, October 21, 2013. |
| 10. |
A chart of historical data can be found at http://www.sca.isr.umich.edu/get-chart.php?y=2013&m=4&n=1r&f=pdf&k=82a71560c431810d7c5948f06a816a11. See also Ben Schnekel, "Consumer Sentiment in U.S. Fell to 10-Month Low in October, Bloomberg, October 25, 2013. The 2011 debt limit impasse is described in CRS Report RL31967, The Debt Limit: History and Recent Increases, by [author name scrubbed]. |
| 11. |
Jeffrey Jones, "U.S. Economic Confidence Continues to Slide Amid Shutdown," Gallup Economy, October 15, 2013, http://www.gallup.com/poll/165431/economic-confidence-continues-slide-amid-shutdown.aspx. |
| 12. |
Conference Board, "Consumer Confidence Decreases Sharply in October," press release, October 29, 2013, http://www.conference-board.org/data/consumerconfidence.cfm. |
| 13. |
Mark Zandi, "U.S. Macro Outlook: Back to the Brink," Moody's Analytics, October 7, 2013, p. 2. |
| 14. |
Government Accountability Office, 2013 Government Shutdown, GAO-15-86, October 2014, p. 38, http://www.gao.gov/assets/670/666526.pdf. |
| 15. |
Many forecasters, including CBO, assume that multiplier effects are smaller when the economy is at full employment and larger when it is below full employment. See Felix Reichling and Charles Whalen, Assessing the Short-Term Effects on Output of Changes in Federal Fiscal Policies, Congressional Budget Office, Working Paper no. 2012-08, May 2012, p. 3. |
| 16. |
Mark Zandi, "U.S. Macro Outlook: Back to the Brink," Moody's Analytics, October 7, 2013. |
| 17. |
5 U.S.C. 7511 defines furlough as "the placing of an employee in a temporary status without duties and pay because of lack of work or funds or other nondisciplinary reasons." |
| 18. |
Executive Office of the President, Impacts and Costs of the October 2013 Federal Government Shutdown, November 2013, https://www.whitehouse.gov/sites/default/files/omb/reports/impacts-and-costs-of-october-2013-federal-government-shutdown-report.pdf. |
| 19. |
Jim Garamone, "Shutdown Over, But Effects on DOD Continue, Hagel Says," American Forces Press Service, U.S. Department of Defense, October 17, 2013. |
| 20. |
Examples are given in Toluse Olorunnipa et al., "Shutdown Hit Boehner's Favorite Diner as $24 Billion Lost," Bloomberg, October 17, 2013. |
| 21. |
Bureau of Labor Statistics, "Federal Government Shutdown: October 2013 Employment Situation News Release," October 30, 2013, http://www.bls.gov/bls/shutdown_2013_cps_ces.htm. |
| 22. |
Complicating the comparison, data are not seasonally adjusted between September and October. |
| 23. |
Bureau of Labor Statistics, Frequently Asked Questions: The Impact of the Partial Federal Government Shutdown on the Employment Situation for October 2013, November 8, 2013, http://www.bls.gov/bls/shutdown_2013_empsit_qa.pdf. |
| 24. |
The unemployment rate is not based on unemployment claims data. |
| 25. |
Overall initial claims are reported on a seasonally adjusted basis. Initial claims by federal workers are reported only on an unadjusted basis. Data are reported by the Department of Labor at http://www.dol.gov/opa/media/press/eta/ui/current.htm. |
| 26. |
Government Accountability Office, 2013 Government Shutdown, GAO-15-86, October 2014, p. 47, http://www.gao.gov/assets/670/666526.pdf. |
| 27. |
Blue Chip, Economic Indicators, vol. 38, no. 10 (October 10, 2013), p. 1. |
| 28. |
Standard & Poor's, Impact of the Debt Ceiling Debate on the U.S. Economy—Getting Worse by the Day, October 16, 2013, p. 1. |
| 29. |
Blue Chip, Economic Indicators, vol. 38, no. 10 (October 10, 2013), p. 16. |
| 30. |
Ibid. |
| 31. |
Macroeconomic Advisers, "Showdown over a Shutdown," September 25, 2013. |
| 32. |
Goldman Sachs, "Shutdown Continues but Recent Actions Are Likely to Reduce Economic Effects," U.S. Daily Newsletter, October 6, 2013. |
| 33. |
Alec Phillips, "Updated Thoughts on the Federal Shutdown and Its Implications," Goldman Sachs, U.S. Daily Newsletter, October 6, 2013. |
| 34. |
IHS Global Insight, U.S. Economic Outlook, October 2013, p. 5. |
| 35. |
Quoted in Joe Wiesenthal, "Morgan Stanley: Here's How Much the Economy Will Suffer Every Week of a Government Shutdown," Business Insider, September 25, 2013. |
| 36. |
Blue Chip, Economic Indicators, vol. 38, no. 10 (October 10, 2013), p. 16. |
| 37. |
Council of Economic Advisers, Economic Activity During the Government Shutdown and Debt Limit Brinksmanship, October 2013. |
| 38. |
Mark Zandi, "A Budget Battle Postmortem," Moody's Analytics, Dismal Scientist, October 21, 2013. |
| 39. |
Quoted in Michelle Jamrisko, Ben Schenkel, and Catarina Saraiva, "Economists Cut Fourth-Quarter U.S. Growth Forecast on Shutdown," Bloomberg, November 1, 2013. |