A new estimate by the Heritage Foundation that the Senate’s bipartisan comprehensive immigration reform proposal would cost $6.3 trillion in the long run is making waves today. (See a summary by The Washington Examiner’s Susan Ferrechio and the full report here.)

Given how contentious the immigration issue is, it’s no surprise that the report is causing controversy (in fact, critics already prebutted the study based on a similar analysis Heritage did of the 2007 immigration effort.) Over at the blog of the libertarian Cato Institute, Alex Nowrasteh argues, among other things, that a proper accounting of the net cost of immigration legislation should take into account factors such as economic growth generated by comprehensive reform.

For all the criticism of the report, however, I think it does one important thing by making an attempt to measure the cost of the legislation over time. Under the current design of the Gang of Eight Senate proposal, immigrants currently here illegally could become legalized after certain near-term conditions are met, but they can’t become citizens for at least 13 years, at which point they’d start becoming eligible for government benefits. The problem is the Congressional Budget Office only measures the fiscal impact of legislation over a 10-year budget window. Any estimate from the CBO, therefore, won’t be able to take into account a spike in the cost of government programs such as Medicaid or Obamacare’s health insurance exchanges as a result of newly-naturalized immigrants claiming benefits.

So, whether or not one agrees with the methodology of this Heritage study, to be useful, any analysis of the impact of immigration legislation should try to estimate costs as new citizens become eligible for government benefits. Costs in year 13 are still costs, even if they don’t show up in CBO’s projections.