Report | Budget Taxes and Public Investment

Cheaper Than You Think—Why Smart Efforts to Spur Jobs Cost Less Than Advertised

Policy Memo #165

 Download PDF
 Press release

Share this page:

The Local Jobs for America Act (LJAA), H.R. 4812, introduced by Rep. George Miller, would create or save more than 675,000 local community jobs.1 The Act authorizes the expenditure of $75 billion over two years in grants to retain jobs in local government that would otherwise be lost due to budget constraints and to create new jobs in local government employment and non-profit service providers in communities.

Unfortunately, concern over federal budget deficits has become a prime stumbling block to enacting efforts like LJAA that would stimulate job creation. Indeed, rather than examine the effects of legislation through multiple lenses—number of jobs, distributional effect across incomes or geographic regions, the upward or downward pressure on poverty—the deficit impact has been elevated to the most important measure of legislation. Even then, it is often misrepresented: spending and cost are not the same things. This Policy Memo points out that while the $75 billion in outlays from LJAA represents the gross cost of the legislation, a truer accounting of its overall effect on deficits would consider offsetting effects, such as higher tax collections and reduced need for social spending. A reasonable calculation of these offsetting benefits shows that almost $40 billion of the bill’s cost would likely be offset by a combination of higher tax collections and reduced safety net expenditures.

Deficit offsets: The “top-down” approach

The simplest accounting for offsets can be obtained by using the data from the Congressional Budget Office (CBO) on the cyclical component of budget deficits. As an economy currently operating below potential moves closer to full employment, tax collections will rise and safety net spending will fall, leading to lower budget deficits. From 1969 to 2008, each dollar increase in actual gross domestic product (GDP) relative to potential GDP has been associated with a $0.37 reduction in budget deficits (Figure A).

[Figure A]


History shows that enacting the LJAA in an economy characterized by 9.9% unemployment will clearly boost GDP growth. A reasonable estimate is that each dollar in state and local government spending will boost overall GDP by $1.40 in the current economic environment. Applying this multiplier to LJAA spending implies that the $75 billion will translate into a GDP gain of $105 billion. Applying the CBO data on how much $105 billion will reduce deficits implies that this $105 billion in extra GDP will lower the deficit by roughly $39 billion.

This “top-down” estimate of the deficit offsets of LJAA is illustrated below with some simple calculations on savings from particular programs.

Jobs preserved and created by the LJAA:
The “bottom-up” approach to deficit offsets

The Education and Labor Committee staff has provided estimates of jobs that would be retained and new jobs that would be created with the passage of LJAA. We can use these to estimate budget offsets provided by the LJAA’s effect on unemployment insurance spending, personal income taxes, and payroll taxes. The committee’s job estimates are:

Retained workers
204,706 in local government
90,501 in rural area local government

New jobs funded by grants to local government
122,823 in local government
142,782 in community non-profits

New jobs in rural areas
53,215 in local government
63,193 in community non-profits

In addition to the direct job creation numbers supplied by the committee staff, we can estimate indirect job creation by using the same macroeconomic multiplier referenced above: every $1 spent in state and local government generates $1.40 of economic activity. Assuming that the ratio of induced jobs is equal to the ratio of induced spending, this would mean that the roughly 680,000 jobs directly created by LJAA would also support 195,000 indirect jobs. If we instead assume that the direct jobs supported by LJAA are more labor-intensive than average, then this might lower the estimates of indirect job creation. Government jobs do tend to be roughly 30% more labor-intensive than the economy-wide average, so to be conservative we will reduce the estimate of indirect jobs by 30%, using an estimate of 150,000 indirect jobs supported by LJAA.

Indirect jobs

How to calculate offsets stemming from retained workers
Retained workers are those whose jobs would have been lost but are able to remain on payrolls for the duration of the bill’s spending impact (two full years). As a result of their continued employment, these workers will continue to pay federal income and payroll taxes and will not make claims on unemployment insurance.

Income and payroll tax offsets
Estimating the income tax offsets will require some crude assumptions, as income tax varies dramatically based on filing status. At $43,385.602—the average wage for a full-time, full-year government worker—married couples with children would likely be receiving a tax refund, while single filers could be paying 6% or more in taxes. To avoid making assumptions about family composition, we just use the average percentage of income paid in federal income tax for tax payers with income between $40,000 and $50,000. The Tax Policy Center (unpublished data) estimates that this is 2.5%. Table 1 shows that the estimated income tax offset would be $669,907,000.

[Table 1]

Payroll taxes are more straightforward. The gross tax rate for every worker (combined rate of employee and employer contribution) is 15.3%: 12.4% Social Security and 2.9% Medicare. This would amount to a payroll tax offset of $4,099,833,000 (see Table 1).

Unemployment insurance offsets
Unemployment insurance (UI) is paid to workers who lost their job through no fault of their own. According to program statistics from the Department of Labor, the average wage replacement is 47%, and the average number of weeks of benefit collection is 19.79. To estimate the share of these workers who would have claimed unemployment insurance benefits in the absence of LJAA, we need to assume a recipiency rate.

The overall average recipiency rate (40%) is clearly inappropriate. This rate measures recipiency among all unemployed workers. Given that many unemployed workers are ineligible for benefits, whereas almost all laid-off government workers will be eligible, the overall rate is clearly too low. Given the high levels of likely eligibility of these employees, estimates of take-up rates of UI of over 80% for the population at large, and evidence that take-up rates rise robustly as generosity is increased (as occurred with the passage of ARRA), we think that a recipiency rate of 80% is a reasonable assumption. This would result in an estimated UI offset of $1,917,234,000 (see Table 1).

Offsets from new workers
New workers are new hires funded by the bill for two full years of work. Over that time period, they will pa
y federal income and payroll taxes. This offset is calculated the same as for retained workers. While retained workers save money by not making unemployment insurance claims, new workers have the opposite effect—they save money by coming off unemployment insurance, or by leaving a job and opening a position for someone else to come off unemployment insurance. We calculated the effect of 480,013 new and indirect workers on unemployment insurance by using program statistics and prior assumptions—average wage replacement (47%), recipiency rate (40%), average benefit collection (19.79 weeks), and the average weekly wage of all employees ($763.98). This would produce an offset of $7,258,123,000 as a result of these new hires (see Table 1).

Offsets from indirect workers
Indirect workers fill jobs created in the economy because of the economic activity generated by the LJAA. We do not know precisely what industries these jobs will be in, or what the workers’ wages will be, so we use average weekly earnings of all employees, $763.98, according to the Current Employment Statistics, and assume full-year work. These workers will pay federal income and payroll taxes. This offset is calculated in the same manner as for retained and new workers. In terms of unemployment insurance offsets, we treat this group the same as new workers—they will be more likely to claim unemployment insurance benefits in the absence of the LJAA. The offset produced by the creation of these “indirect” jobs is estimated to be $1,664,550,000 (see Table 1).

Evidence on total offsets from tax collections and unemployment insurance

Table 1 summarizes the offsets for retained, new, and indirect employment stemming from increased tax collections and lower spending on unemployment insurance, all based on the assumptions discussed above. The bottom-line: federal income tax collections will be $1.7 billion higher with LJAA than without it; payroll tax collections will be $10.6 billion higher with LJAA than without it; and unemployment insurance spending will be $3.3 billion lower with LJAA than without it.

These numbers illustrate how legislation like the LJAA, enacted during times of high unemployment, can actually be partially self-financing—the jobs and incomes it creates translate into higher tax collections and reduced spending. To be clear, the LJAA will add to the deficit, but by a much smaller amount than the headline gross price tag would suggest. Furthermore, much of the economic activity generated by LJAA is not captured in these jobs-based considerations of deficit offsets—for example, finding office space for government workers requires rental payments that would otherwise not accrue to landlords. It is also worth noting that the indirect jobs generated are likely to be private-sector jobs, so business income and profits will rise. And there are other government safety net programs (Medicaid, Temporary Aid to Needy Families) that will see savings as people retain jobs and find new ones.

In short, the best estimate of the overall offsets contained in LJAA remains the CBO estimates of the effects of GDP growth on budget deficits. The logic of these numbers is illustrated with the examples we have presented on tax collections and unemployment insurance savings stemming solely from improved labor market conditions.


Given the pervasive concern over federal budget deficits, the debate over job creation has too often ignored the net cost of action. The original Recovery Act spurred income creation that resulted in higher tax collections and lower safety-net spending, substantially blunting its bottom-line impact on deficits. While the LJAA does indeed spend serious money aimed at creating jobs in the short-run, its net cost will be much lower than advertised as it puts people back to work and turns them into tax-payers rather than benefit-collectors.

The undue emphasis placed on a given piece of legislation’s deficit impact, however, often overshadows the actual problems the legislation is trying to address. Indeed, rather than examine the impact of legislation through multiple lenses—number of jobs created, distributional effect across incomes or geographic regions, the upward or downward pressure on poverty—the effect on the deficit has been elevated to the most important measure of legislation. The LJAA would take measures to spur job growth, but they are not nearly enough. Coming out of the worst recession since the Great Depression, it is job growth, not deficits, that should be our bottom line.


1. Estimates of job creation provided by House Education and Labor Committee staff.

2. Annual salary of a full-time, full-year local government employee is estimated using the average hourly wages of local government employees in 2009, obtained from the Outgoing Rotation Group of the Current Population Survey (or CPS-ORG).

See related work on Budget Taxes and Public Investment

See more work by Josh Bivens and Kathryn Anne Edwards