President’s Budget Proposes Unemployment Insurance Reforms

Today, the Obama administration released its fiscal year 2017 federal budget proposal, including a series of proposals to strengthen and modernize the federal-state unemployment insurance (UI) program. NELP will release more in-depth analyses of key features of the UI and reemployment proposals in the coming days, but in the meantime, we think it’s important to draw attention to the labor market context in which these much-needed recommendations are being made.

Since U.S. employment hit its Great Recession low six years ago, state UI programs have not performed well when measured according to a key indicator known as the recipiency rate.[i] The recipiency rate measures the proportion of jobless workers who get UI benefits.[ii]

In fact, the majority of unemployed people in the United States don’t get UI benefits. NELP called attention to the record-low recipiency rate for calendar year 2014 in its report, The Job Ahead. Using the latest data, we find that the recipiency rate in 2015 remained at a record low, with just over one in four jobless workers (27 percent) receiving UI benefits in 2015. Figure 1 below shows the national recipiency rate from 1972 through 2015, reported as a 12-month moving average.


Several States Offer Little Protection from Unemployment

While the national recipiency rate remained at record lows in 2015, state recipiency rates are widely mixed, ranging from just 11 percent in Florida to 66 percent in North Dakota (see Table 1 at the bottom of this post). With the exception of North Dakota, all state UI programs compensated fewer than half of jobless workers in 2015. The median state, Oklahoma, provided benefits to about 28 percent of its jobless workers. More troubling, 14 states paid benefits to 20 percent or less of their jobless residents.

Four of the five states with the lowest recipiency rates in 2015—Florida (11 percent), South Carolina (12 percent), North Carolina (13 percent), and Georgia (13 percent)—adopted significant cuts to available weeks of benefits starting in 2011. With the exception of South Carolina (which pays only 20 weeks of UI), these four states adjust their available weeks of benefits on a sliding scale based upon state unemployment rates. Today, someone who loses her job in Florida can only qualify for up to 12 weeks of benefits, less than half the national average duration of unemployment in 2015. (For the number of available weeks of benefits in each state, see this helpful map from our partners at the Center on Budget and Policy Priorities.)

UI is designed as a social insurance program to compensate wage losses, so using recipiency as a performance indicator is one way to see how broadly a state UI program is offering protection to those who are out of work. Since the income replacement and economic impacts of UI are dependent upon paying benefits to a significant proportion of people who are unemployed, low recipiency rates show that state program performance is undercutting the national goals of UI.[iii] Ensuring that state UI programs provide reasonable protections to jobless workers is clearly a priority, and in the next few days we’ll focus on elements of the Obama administration’s budget that will address this need.

Table 1. Percentage of jobless workers receiving unemployment insurance in 2015 (Updated April 2016) [iv]

State 2015 UI Recipiency Rate
Alabama 17%
Alaska 46%
Arizona 15%
Arkansas 30%
California 33%
Colorado 29%
Connecticut 40%
Delaware 31%
District of Columbia 32%
Florida 11%
Georgia 14%
Hawaii 34%
Idaho 25%
Illinois 31%
Indiana 18%
Iowa 38%
Kansas 28%
Kentucky 23%
Louisiana 15%
Maine 30%
Maryland 25%
Massachusetts 43%
Michigan 26%
Minnesota 43%
Mississippi 15%
Missouri 21%
Montana 38%
Nebraska 24%
Nevada 26%
New Hampshire 19%
New Jersey 45%
New Mexico 21%
New York 35%
North Carolina 12%
North Dakota 70%
Ohio 24%
Oklahoma 28%
Oregon 30%
Pennsylvania 45%
Rhode Island 32%
South Carolina 13%
South Dakota 14%
Tennessee 15%
Texas 29%
Utah 21%
Vermont 42%
Virginia 17%
Washington 27%
West Virginia 32%
Wisconsin 36%
Wyoming 45%




[i] Many analysts have used UI recipiency rates when studying state UI programs. See, for example, Wayne Vroman, “Low Benefit Recipiency in State Unemployment Insurance Programs,” ETA Occasional Paper 2002–02, U.S. Department of Labor (June 2001), available at David Wittenberg, et al. provided an extensive review of the recipiency studies to date in “Literature Review and Empirical Analysis of Unemployment Insurance Recipiency Ratios,” Final Report to U.S. Department of Labor (June 1999), available here. Gary Burtless first called attention to falling recipiency rates in 1984. See Gary Burtless and Daniel H. Saks, “The Decline in Insured Unemployment During the 1980s,” Brookings Institution for the U.S. Department of Labor (1984).

[ii] There are several ways to measure UI recipiency. Consistent with prior practice, NELP is calculating recipiency by comparing, for each state, the average weekly number of continued weeks claimed—also known as the number of insured unemployed—with the average monthly number of unemployed people. The first figure is based on data reported by state UI agencies to the U.S. Department of Labor; the second figure is based on monthly estimates from the Current Population Survey, the primary source of national labor force statistics. Others calculate recipiency rates using weeks compensated relative to the number of totally unemployed. Weeks compensated exclude weeks for which claimants are serving penalty or waiting weeks, so better reflect the number of people actually receiving UI payments from week to week. And, some have calculated a recipiency rate for job losers (those most presumptively eligible for UI) or for the short-term unemployed (since state programs usually pay benefits for 26 weeks). Regardless of the method used, all recipiency rate measures move in similar patterns and show that state UI programs paid benefits to fewer jobless workers in recent years than in any prior years since World War II, exceeding even the previous record-low years of the early-1980s. These measurement issues are discussed in more detail in Will Kimble and Rick McHugh, “How Low Can We Go? State Unemployment Insurance Programs Exclude Record Numbers of Jobless Workers,” Economic Policy Institute (March 2015), available at (see notes 2 through 6 and accompanying text) and Stephen A. Wandner and Thomas Stengle, “Unemployment Insurance: Measuring Who Receives It,” Monthly Labor Review (July 1997), p. 15, available at

[iii] Wayne Vroman, “The Role of Unemployment Insurance as an Automatic Stabilizer,” Report ETAOP 2010-10, Impaq International for the U.S. Department of Labor (2010), available at; Wandner and Stengle, op. cit; and U.S. General Accountability Office, “Declining UI Recipiency,” GAO/HRD-92-34R (April 28, 1992), available at

[iv] NELP calculations of monthly UI continued weeks claimed data, from ETA report 5159, Claims and Payment Activities, U.S. Department of Labor, available at Regular programs data for December 2015 provided upon request. State unemployment data are from the Bureau of Labor Statistics. Regular programs include State UI, UCFE, and UCX. Continued weeks claimed include waiting and penalty weeks.

Back to Top of Page