Examining the Congressional Budget Office Cost Estimate of the Senate Immigration Bill
July 15, 2013
The Congressional Budget Office (CBO) on July 3 issued its estimate of the budgetary impact of the final Senate-passed immigration reform bill, as amended when the bill was considered on the Senate floor. (CBO had issued a prior estimate based on the version of the bill reported by the Senate Judiciary Committee.). CBO estimates that the bill would significantly reduce deficits over each of the next two decades — by $158 billion in the first decade, fiscal years 2014-2023, and by about $685 billion over the second decade — because higher tax payments from immigrants would exceed the increased costs resulting from the bill, such as for federal benefits for some legal immigrants and increased border security.
The Senate bill would expand the economy by increasing the population and labor force, CBO found, thereby boosting gross domestic product (GDP) by approximately 3.3 percent in 2023 and approximately 5.4 percent in 2033. (These percentages are the midpoints in range projections that CBO provides.) Average wages would grow slightly more slowly in the short term than they otherwise would, as the economy adjusted to the larger workforce. This partly reflects the fact that the immigrants who would come in under the bill will have lower wages, on average, than current residents, and does not necessarily imply that current residents would be worse off. By 2025, however, the bill would raise average wages above what they would otherwise be across the wage distribution.
Bill Would Reduce Deficits by Roughly $160 Billion in First Decade
CBO estimates that the bill would reduce deficits by a total of $158 billion over 2014-2023. This figure represents the bill’s net impact on mandatory spending and revenues. (When CBO estimated the budgetary impacts of the bill reported by the Judiciary Committee, it found that the bill would have reduced deficits by $197 billion — the difference stems from an amendment that added significant new funding for border security-related measures.)
CBO also estimates that implementing the bill would raise discretionaryprogram costs by $23 billion over this period. If Congress increased the 2011 Budget Control Act’s (BCA) caps on discretionary program funding by this amount, this would shrink the bill’s deficit reduction to $135 billion. On the other hand, if Congress funded these additional costs within the existing BCA caps, the increased discretionary funding to implement the bill would have no net effect on total discretionary spending or deficits over the ten-year period.
The CBO report shows the bill’s effect on mandatory spending and revenues in three ways: on total budget deficits, on “on-budget” deficits, and on “off-budget” deficits. The “on-budget” analysis includes all mandatory spending and revenues except Social Security spending and Social Security payroll taxes; the “off-budget” analysis covers only Social Security. The total budget analysis includes both “on” and “off” budget pieces.
- The overall impact, as noted, would be to reduce total deficits by $158 billion over ten years.
- The bill would modestly increase the “on-budget” deficit, raising it by $52 billion over 10 years.
- In contrast, the bill reduces the “off-budget” deficit by $210 billion over the same period.
While some may say that the proper way to analyze the bill is to look only at the “on-budget” deficit estimate, this ignores the bill’s significant improvement to Social Security’s near-term finances. As the Social Security actuary, Stephen Goss, wrote Senator Marco Rubio (R-FL) in May, the bill would improve Social Security’s short- and long-term solvency.
In the second decade following enactment of the bill (2024-2033), CBO estimates that it would reduce deficits by a total of about $685 billion. (CBO notes that this estimate is uncertain.) CBO estimates that $110 billion of this total deficit reduction will be “on-budget,” with the remaining $575 billion coming from “off-budget” improvements to Social Security’s finances.
Over the next 20 years, CBO’s best estimate of the bill’s overall impact is to reduce total deficits by about $840 billion. The bill significantly reduces deficits over the next two decades because the increased taxes paid by immigrants — both new entrants to the country under the bill’s changes in legal immigration policies and workers already in the country who go through the steps needed to convert from undocumented to legal status — would exceed the increased costs resulting from the bill.
Those increased costs include providing some federal benefits to legal immigrants and their children, and increased border security and other spending related to immigration enforcement. (Fees that immigrants would pay would also help to offset some of the bill’s costs.)
A large share of the bill’s effects on both revenues and expenditures would come from the increase in the number of legal immigrants allowed to enter the United States as a result of its changes to immigration policies.
How CBO Accounted for the Bill’s Economic Effects
The bill would expand the size of the U.S. population and workforce, which in turn would directly increase the size of the economy, the revenue base, and spending for federal benefit programs. CBO took these direct effects into account in its official cost estimate of the bill’s budgetary impact.
CBO has a long-standing practice of using, to the maximum extent possible, the same assumptions about broad macroeconomic variables like GDP and employment in its cost estimates that it uses in its baseline budget projections. Therefore, the official cost estimate for the immigration bill does not reflect a broader set of interactions between the budget and the economy (“macroeconomic feedback”), including changes to business investment and productivity resulting from the larger population and labor force. CBO did not, for instance, include in the cost estimate effects on GDP over and above the direct increase stemming from the increase in the population and the labor force. Thus, the cost estimate does not reflect how any “dynamic” change in GDP would affect revenues.
There are very sound reasons why CBO and the Joint Committee on Taxation do not use such “dynamic scoring” in their cost estimates. Economists do not agree on the size — or sometimes even the direction — of macroeconomic feedback from reducing marginal income tax rates or other changes to tax laws or program provisions. Dynamic scoring would also impair the credibility of the budget process: because the estimates of macroeconomic feedbacks are very uncertain, including them in budget estimates would be highly controversial and inevitably viewed as biased and politically motivated.
Nevertheless, as it sometimes does with major legislation, CBO provided a separate analysis of the bill (as reported by the Senate Judiciary Committee) with a range of estimates of additional economic effects not in the official cost estimate. CBO did not issue a revised analysis of these economic effects on the version of the bill that passed the Senate. However, in its July 3rd cost estimate, it indicated that “Under the version that was passed by the Senate, those broader effects would differ only slightly from those estimated for the earlier version of S. 744.”
CBO’s economic analysis finds that the economic effects not incorporated into the official cost estimate “would have no significant effect on federal budget deficits during the coming decade and would reduce deficits during the following decade” (emphasis added). CBO estimates that these broader economic effects would reduce deficits by about an additional $300 billion in the second decade, beyond the $685 billion in direct effects, based on the midpoint of its estimated range of economic effects. CBO also noted, however, that the effects of the legislation “on the economy and the federal budget would be complicated and highly uncertain, even in the short run, and the actual effects could be well outside CBO’s ranges of estimates.”
Bill Would Expand Economy and Eventually Raise Average Wages
In broad terms, CBO finds that the increase in the population and the labor force resulting from the bill would raise GDP. Initially, the bill would modestly reduce average wages and per-capita income (compared with what they would be without the bill). Over time, however, as businesses expanded to absorb the increase in the labor force and to meet the additional demand from a larger population more efficiently, the economy would be larger, and average wages and productivity would be higher, than without the bill.
CBO estimates that the bill would boost GDP by 3.3 percent in 2023 and by 5.4 percent in 2033, relative to what GDP would be in the absence of the legislation. This increase stems from several factors, including an increase in the labor force, increased capital investment, and increased productivity. CBO projects that both overall GDP and average per-capita GDP would be higher in 2033 as a result of the bill.
CBO projects that the influx of new labor would slightly reduce average wage growth over the first decade. The level of average wages in 2023 would be about 0.1 percent lower than it would be under current law. CBO indicates, however, that this does not necessarily mean that average wages would be lower than they otherwise would be for current U.S. residents. At least part of the effect on wages would come from the increase in new immigrants, who earn lower wages than current residents, on average. CBO did not separately analyze the wage effects for those who would reside in the United States in the absence of the bill, and there is no basis in the CBO report for concluding that the bill reduces average wages for current U.S. residents.
By 2025, the bill would raise average wages across the wage distribution, according to CBO, and average wages in 2033 would be 0.5 percent higher than they otherwise would have been. Importantly, CBO estimates that average wages in 2033 would be higher for workers “in all quintiles of the skill distribution.” That is, low-, medium-, and high-skilled workers would all see higher average wages as a result of the bill by 2033.
CBO estimates that, under the bill, the unemployment rate for the next five years would be “roughly 0.1 percentage point higher relative to projections under current law.” However, CBO estimates no increase in the unemployment rate over the longer run as a result of the bill. As with wages, CBO does not distinguish between the unemployment rate for existing residents and the rate for new immigrants, so there is no basis in the CBO report for concluding that the bill increases unemployment for current U.S. residents.
In addition to the economic growth flowing from increased business investment in machines, factories, and stores to meet the expanded demand for goods and services from a larger population and to make the best use of the larger labor force, CBO estimates that the increased immigration — particularly of higher-skilled immigrants — would “tend to generate additional technological advancements, such as inventions and improvements in production processes” that would make workers and capital more productive, raising wages and incomes further.
CBO makes clear that the estimates of these economic impacts and the resulting budgetary impacts are uncertain. For each key variable provided, it gives both a single “central tendency” value and a range of plausible estimates.
Higher Tax Revenues Would More Than Offset Higher Spending
CBO estimates that the bill would raise overall revenues by $456 billion over 2014-2023. Nearly all (98 percent) of that would come from increased income, Social Security, and Medicare taxes. The rest would come from unemployment taxes and various fees and penalties in the bill.
Over the same period, overall “direct” spending (that is, spending on mandatory programs) would grow by $298 billion as a result of the bill.
Most of the increased revenues and spending would result from an increase in the number of people living in the United States who work, pay taxes, and sometimes qualify for benefits such as Medicaid, Medicare, or unemployment benefits.
Over the ten-year period, the two areas with the largest cost increases would be subsidies for some lawful permanent residents and other legal workers to buy health insurance through the state insurance exchanges and the cost of refundable tax credits. There is also significant funding for border security. But increased tax revenue from immigrants would more than offset the costs of these and other federal benefits, over both the next ten years and the subsequent decade.
The bill includes substantial benefit restrictions for immigrants converting from undocumented to legal status. Immigrants who convert to registered provisional immigrant (RPI) or blue-card status — the two “legalization” categories that would apply to currently unauthorized residents and certain agricultural workers, respectively — would be ineligible for SNAP (food stamps), Medicaid, and subsidies to buy health insurance through the exchanges. Most also would remain ineligible for key benefits such as SNAP and Medicaid for five years after they attain lawful permanent resident (LPR) status. Thus, a large share of the increased costs of these benefits would not be for immigrants who have converted from undocumented status, but rather for immigrants who entered the country legally and would eventually qualify for assistance.
 While CBO did not revise its economic analysis of the bill after passage in the Senate, in its updated budget analysis, CBO notes that the economic effects of the later version of the bill would only differ slightly from those of the previous committee-reported version. (CBO cost estimate as passed, p. 2)
 Note that the analysis of budgetary impacts was released by CBO, but the estimates were done by both CBO and the Joint Committee on Taxation (JCT), which is charged with estimating the impact of legislation on revenues. When this document refers to CBO estimates or projections, it refers to estimates by both CBO and JCT.
 Letter from Stephen C. Goss, Chief Actuary, to the Honorable Marco Rubio, May 8, 2013, http://www.ssa.gov/OACT/solvency/MRubio_20130508.pdf.
 CBO cost estimate, July 3, p. 2.
 These are “central estimates” of the full range of CBO’s estimated effects. Under the full range, CBO estimates that the bill could boost GDP by between 5.1 and 5.7 percent in 2033, although the agency acknowledges considerable uncertainty in these estimates and that the actual effects could be “well outside” this range.
 In its prior estimate, CBO indicated that the language that bars immigrants in RPI and blue-card status from federal means-tested benefits could be interpreted to permit some immigrants in these statuses to receive benefits in certain circumstances. Specifically, CBO’s estimates assume that some children and pregnant women in RPI or blue-card status would be eligible for Medicaid or CHIP (both at state option) and that some children in these statuses would be eligible for SNAP. The costs that CBO estimates in these programs as a result of this interpretation appear to be very modest, however, because most children living with undocumented parents are U.S. citizens, not undocumented immigrants, and thus are unaffected by the section of the legislation that’s in question.