The Congressional Budget Office (CBO) recently released its long-awaited report on the likely budgetary effects of the American Health Care Act. The legislative counterpart to the White House’s Office of Management and Budget, CBO estimates how federal spending and revenues would change as a result of proposed legislative bills. The resulting Republican talking points were familiar: the health care bill’s numbers were biased, they protested, predicting the future is hard. They were familiar not only because they have become Trump’s refrains du jour for “so-called” expertise (think intelligence agency reports or unemployment statistics), but also because they are the same critiques often lobbed at another little-known office that wields outsized influence: the Office of Information and Regulatory Affairs (OIRA).

Among other things, OIRA is charged with reviewing the cost-benefit analyses (CBAs) of executive agencies’ significant regulatory actions. Its mission is to ensure that “the benefits of the intended regulation justify its costs.” A more recent Trump executive order also has the office scrutinizing the “total incremental costs” of new regulations to ensure they remain within fixed regulatory budgets.

Like the CBO Director, the OIRA Administrator is often a punching bag for both the Left and the Right. When you’re trying to maintain a reputation for nonpartisan number-crunching, you can’t please everyone. And like CBO, OIRA must also make predictions about the future amidst uncertainty: How many companies will go out of business as a result of technology-forcing requirements? How many people will no longer get lung cancer as a result of tobacco warning labels? These judgments require answers to hard questions about the right modeling assumptions, discount rates, and time horizons.

One striking aspect of the recent CBO controversy has been how much it shows the potential for participants in both processes, CBO and CBA, to learn from each other. (As someone who worked at OIRA, I know more about the latter than the former, so will pitch some of the analogous questions to the CBO experts out there.) Sure, there are analytical differences between budgetary and regulatory costs, but at a higher level of generality, note that the CBO-CBA analogy becomes more acute in light of the aforementioned need for OIRA to consider regulatory as opposed to fiscal “budgets.” Here are three areas of potential overlap, but there are surely more—perhaps the topic of future posts: 

• Retrospective review. One thing we learned in light of Press Secretary Sean Spicer’s accusation that CBO was “way, way, off the last time in every aspect of how they scored and projected Obamacare,” was that CBO learns too. CBO, that is, updates its numbers to reflect new information. So while it initially projected that the ranks of the uninsured would drop by 30 million under the Affordable Care Act, six years later, CBO reduced its estimate to 22 million in light of actual, observed behavioral changes.

Similarly, many commenters have called for agencies to update their projected costs and benefits in light of new data. An Obama executive order enshrined that demand by requiring agencies to consider what is known in the CBA-world as “retrospective review”: looking back at initial cost-benefit estimates and revisiting them. More recently under Trump’s new order, OIRA has stated that agencies have to determine “the most current information available on projected cost savings” instead of blindly relying on its original estimates. The point is that agencies will now have to take retrospective review much more seriously.

CBO apparently has an internal process and timetable in place for revisiting its original CBO scores. How does it work and to what extent can and should other agencies conducting CBA mimic it? At bottom, these updating requirements force government actors to acknowledge their errors—something that might cause short-term embarrassment, but that will ultimately enhance their long-term credibility.

• Macroeconomic effects. In May 2015, Congress adopted new budget rules requiring CBO to incorporate macroeconomic effects into its 10-year cost estimates for major legislation. Incorporating macroeconomic feedback effects into such estimates is known as “dynamic scoring.” The practice is controversial in part for introducing even more uncertainty into an already tentative set of estimates. As a result, CBO has gamely issued a number of reports and thoughtful analyses on how to incorporate projected budgetary effects on labor supply and the fiscal multiplier. (Interestingly, its recent health care bill analysis candidly admits that CBO didn’t have sufficient time to estimate the bill’s macroeconomic effects).

The CBA-crowd has also grappled with the possibility that cost-benefit analyses should incorporate macroeconomic effects. Eric Posner and Jonathan Masur, for example, have asked whether regulations should be countercyclical. Cary Coglianese, Adam M. Finkel, and Christopher Carrigan have a thoughtful edited volume exploring the relationship between regulations and employment. One suspects there is much more to be learned from the CBO, which has been actively operationalizing closely related questions for a group of general policymakers. What have been the conceptual barriers in that context and the political minefields to avoid?

• Qualitative impacts. Finally, and the overlap here is more inchoate, but CBO’s dynamic scoring requirements also call for the office to include a “qualitative” assessment of budgetary effects for 20 years following its initial 10-year outlook. What does this mean in practice? According to CBO:

“A qualitative assessment of effects in later decades might look something like this: Including macroeconomic effects, CBO expects that the legislation would (increase/decrease) the deficit during the 2015–2025 period and would (increase/decrease) it in the following 10 years; those (increases/decreases) in the following 10 years would probably be (larger/smaller) than the ones projected for the 2015–2025 period.”

As a legislator considering a bill, one might wonder what to do with qualitative projections like this: How should they be traded off against the quantified macroeconomic effects? What role should these qualitative estimates play in congressional debates?

And here, finally, is an area in which CBO might learn something from CBA. For decades, executive orders have allowed agencies to consider both “qualitative” costs and benefits. Many thoughtful scholars have thus grappled with analogous questions such as why agencies choose to present qualitative figures, how they should do so, and whether unquantifiable effects compromise the entire cost-benefit enterprise.

For all these reasons, getting CBO and CBA wonks in the same room, so to speak, is likely to yield rich dividends. Admittedly, that conversation may have to occur later rather than sooner: Trump’s legislative and (de)regulatory agenda is keeping both camps plenty busy.

Taking a step back in the meantime, CBO scores and CBA ultimately implicate tougher questions about expertise versus politics, whether boundaries exist and, if so, who should police them. This comparison also promises insights on how Congress and the President seemingly bind themselves to certain principles in advance—and why they later renege on them. These themes are hardly new to the administrative state, but the Trump Administration once again underscores just how high their stakes can be.

[Cross-posted at Take Care.]