Wall Street Journal columnist Caril Bialik notes that estimates by the Congressional Budget Office of the effects of proposed legislation are likely to be highly inaccurate. He notes several reasons for this, including rules that require CBO to rely on assumptions that everyone knows are false:
The agency is in a tough spot. Its purpose is to give Congress a quick take on how a law would affect the nation’s finances, and occasionally to assess broader economic impact. But, unlike private forecasters, it is constrained by rules on what can and can’t be taken into account. For instance, the CBO assumes existing laws will prevail, or expire as scheduled, even if the political reality likely is different.
Another reason why CBO estimates are unreliable is the CBO refuses to discount future effects, something that is taught to every student enrolled in a course on benefit-cost analysis. In the case of the immigration bill recently passed by the Senate, CBO has company from at least one Washington-based think tank: the Heritage Foundation. Unlike the CBO, however, Heritage economists are not compelled by legislators to rely on substandard and misleading methods.
In support of its campaign to defeat the Senate-passed legislation, Heritage published a report by Robert Rector and Jason Richwine estimating the net fiscal costs imposed by “unlawful immigrants.” This is not the same as a benefit-cost analysis. It counts only benefits relived from government programs net of taxes paid. Benefit-cost analysis would count net fiscal costs (or net fiscal benefits) as transfers, not as social costs (or benefits).
Rector and Richwine estimate that under current law unlawful immigrants impose an aggregate annual fiscal deficit of $54.5 billion. Legislation similar to the Senate bill would reduce this fiscal deficit to $43.4 billion per year during th interim phase before previously unlawful immigrants would become eligible for citizenship, after which the aggregate fiscal deficit would rise to $106 billion per year. Once eligible for Social Security and Medicare, Rector and Richwine estimate the aggregate annual fiscal deficit at around $160 billion per year.
They add these figures to obtain a summary estimate, which has been widely quoted:
Over a lifetime, the former unlawful immigrants together would receive $9.4 trillion in government benefits and services and pay $3.1 trillion in taxes. They would generate a lifetime fiscal deficit (total benefits minus total taxes) of $6.3 trillion. (All figures are in constant 2010 dollars.) This should be considered a minimum estimate. It probably understates real future costs because it undercounts the number of unlawful immigrants and dependents who will actually receive amnesty and underestimates significantly the future growth in welfare and medical benefits. (p. vii)
Rector and Richwine control for inflation but they do not discount. That is, they give the same weight to fiscal costs regardless of when such costs are realized. This is not explained anywhere in the report. Bialik interviewed Rector for his column and notes that the absence of discounting afflicts both the Rector-Richwine study and the CBO estimate:
There is one thing Mr. Rector’s report has in common with the CBO’s: His is also tailored to an audience in a way some economists dispute. The $6.3 trillion in spending isn’t discounted to reflect that a dollar today is worth a lot more than a dollar in 50 years. Such discounting could cut half or more from the figure.
Mr. Rector said Heritage typically doesn’t account for so-called net present value of long-term costs and revenue.
“At Heritage, we are not comfortable using net present value because it makes costs seem very small to decision makers,” he said. “For nine out of 10 congressmen, that’s just a smaller number if you do that.”
This is not correct. Discounting also makes future benefits smaller, and in other contexts Heritage has reported approvingly of future effects expressed in discounted terms. It is easy to find examples on Heritage’s blog “The Foundry” — posts on large-scale mortgage forgiveness, solar energy subsidies, government debt, Social Security, and estimates of the benefits and costs of cap-and-trade to control greenhouse gas emissions.
Rector and Richwine would have been on firmer ground, and avoided unnecessary criticism, if they had discounted the future effects of proposed immigration legislation. That CBO does not discount speaks poorly of CBO; even if it is required to report undiscounted figures, it is not prohibited from supplementing them with properly discounted estimates. Rector’s implication that Members of Congress cannot be expected to understand estimates presented in net present value terms speaks poorly of Members of Congress, to be sure, but it disconcertingly suggests that they — and the American people generally, who Rector and Richwine sought to inform — are beyond being educated.
We prefer the term unauthorized aliens because it better reflects the language in the Immigration and Nationalities Act. Sec. 101(a)(3) [8 USC 1101(a)(3)] defines alien as “any person not a citizen or national of the United States.” The term immigrant is defined in Sec. 101(a)(15) to encompass all aliens except those who belong to one or more excluded classes of nonimmigrant aliens (e.g., accredited diplomatic personnel and their attendants, crewmen, businessmen, tourists, and students). Any alien who is not authorized to be present in the United States is an unauthorized alien. We assume that Heritage intends that its term — unlawful immigrants — to be the same as unauthorized aliens.