The Congressional Budget Office, on the fourth floor of the Ford House Office Building in Washington. (Melina Mara/The Washington Post)

Marc Short is assistant to the president for White House legislative affairs. Brian Blase is special assistant to the president for the National Economic Council.

In the coming days, the Congressional Budget Office will release an updated analysis of the Senate bill to repeal and replace Obamacare. The CBO will likely predict lower health insurance coverage rates if the bill becomes law. The American people and Congress should give this prediction little weight in assessing the bill’s merit.

The reason: The CBO’s methodology, which favors mandates over choice and competition, is fundamentally flawed. As a result, its past predictions regarding health-care legislation have not borne much resemblance to reality. Its prediction about the Senate bill is unlikely to fare much better.

When Obamacare passed in 2010, the CBO projected a healthy individual market with 23 million people enrolled in exchange plans by this year. The CBO predicted that by 2017, exchange plans would be profitable and annual premium increases low.

The CBO reached these conclusions, in large part, because its model puts significant weight on the individual mandate. The CBO expected millions of relatively young and healthy people to buy exchange plans under government coercion.

But this never happened. Today, there are only 10 million people enrolled in exchange plans — about 60 percent fewer than expected. (Contrary to some claims, this is not because more people have maintained employer plans than the CBO expected; the reduction in employer coverage has been greater than the CBO projected, and overall about 9 million more people are uninsured now than projected.) Absent the projected bounty of young, healthy consumers, health insurers are abandoning the exchanges, leaving a third of American counties with only one insurer to choose from. As insurers continue to flee the exchanges, consumers will face even fewer options next year.

And while choice is declining, costs are skyrocketing. A recent analysis by the Department of Health and Human Services found that the average annual premium on the individual market has more than doubled since 2013 — up nearly $3,000 in only four years. Premiums seem set to increase at least 25 percent next year as well. These hikes, along with the large drop in insurer participation, show that many state exchanges are descending deeper into adverse-selection spirals.

The CBO failed to foresee any of this. Despite the obvious shortcomings of its previous analyses, the CBO has utterly failed to update its model to account for reality. Instead, the CBO continues to attribute mythical power to the individual mandate.

Two examples clearly show the flaws. First, the CBO believes that about 15 million people value their insurance so little that they will simply drop coverage next year following the repeal of the individual and employer mandates — an unlikely occurrence given the individual mandate’s inability to cause people to enroll thus far. Furthermore, this figure includes approximately 4 million people on Medicaid, even though Medicaid enrollees pay little or nothing for coverage and are largely exempt from the penalty. It makes little sense for fewer individuals to have Medicaid coverage just from the repeal of the individual mandate, yet the CBO predicts millions of fewer Medicaid enrollees.

Second, the CBO estimates that the Obamacare exchanges will average 18 million enrollees next year assuming the law remains in place. Yet only about 10 million Americans had exchange plans in 2015, 2016 and 2017 — and the CBO ignores Obamacare’s collapse. Simply put, the CBO predicts coverage under Obamacare that will never materialize.

The CBO’s belief in the power of the individual mandate skews its premium estimates as well. If the mandate actually compelled millions of additional healthy and young people into the market, as the CBO assumes, lower average premiums would result. But such consumers have largely steered clear of the law. By failing to account for this reality, the CBO’s estimate of any replacement measure assumes that consumers will leave a market they never joined in the first place — and correspondingly estimates higher premiums than will materialize.

Beyond the individual market, the CBO has also failed to properly analyze Obamacare’s Medicaid expansion. For one, the Medicaid expansion to working-age, non-disabled adults has proved to be substantially more expensive than the CBO projected — nearly 50 percent more expensive per enrollee.

The forthcoming analysis of the Senate bill will likely contain additional projection errors. It will almost certainly assume that many additional states would expand Medicaid under Obamacare, even though other agencies, such as the Office of the Actuary at the Centers for Medicare and Medicaid Services, have determined this is highly unlikely. An assumption that states will expand Medicaid means that the CBO will determine that millions of people will “lose” Medicaid coverage under the Senate bill — but these people were never enrolled in the first place.

The American people — and their representatives in Washington — deserve the most accurate assessment possible of the effects of critical health-care legislation. Although the CBO generally plays a valuable role in the legislative process, as Obamacare’s ongoing failure clearly demonstrates, the CBO’s health-care model is fundamentally flawed. The CBO’s failure to update the model means its forthcoming analysis of the Senate bill will be no better — and perhaps worse — than its disproven Obamacare projections. Although the media and the political left will certainly seize on it, the CBO’s estimates will be little more than fake news.