President Trump’s policy agenda is in trouble. After nearly seven months of intramural fighting and several dramatic reversals, the Republican effort to repeal and replace the Affordable Care Act died in the Senate. By the end of last week, even the most stalwart opponents of Obamacare seemed ready to move on.

At the center of the Trump White House’s difficulties has been its own inexperience with the complex web of statutes, organizations and procedures that constitute contemporary American government. Trump’s aversion to the details of dealmaking means that even if congressional Republicans manage to pass a plan to repeal, replace, repair or revise Obamacare, it will be despite — not because of — help from the presidency.

Yet, as I argue in a forthcoming issue of The Forum, it would be a mistake to evaluate Trump’s effectiveness on the basis of legislative failures alone. Whereas other presidents have used policy overhaul — prepared by experts — to change politics, Trump is using the presidency to generate political conflict and to undermine the cadre of policy experts who make government work.

Trump has put experts on ice

The federal bureaucracy presents a formidable challenge for presidents with an aggressive deregulatory agenda like Trump’s. Job protections enable bureaucrats to continue to produce expert analyses that may undermine the case for deregulation.

Trump has responded to this problem with what Walter Williams calls an “anti-analytic” strategy. The administration’s budget has proposed major cuts to programs of policy-relevant scientific research on climate and the environment, as well as a plan to eliminate the Agency for Healthcare Research and Quality, which analyzes the safety and effectiveness of medical treatments.

The administration’s political appointees have also put pressure on outspoken career civil servants. After repeatedly speaking out about the effects of climate change on Native American communities, Joel Clement — the former director of the Interior Department’s Office of Policy Analysis — was involuntarily reassigned to the accounting department. Clement, one of 50 senior department employees receiving such reassignments, has filed a complaint with the U.S. Office of Special Counsel.

Yet the administration’s effort to remove experts from government has gone beyond reassigning troublesome officials. Government agencies depend on advice from scientific advisory boards made up of academic researchers. These boards have been a central target for the administration. In early May, the Environmental Protection Agency fired half of the members of an 18-person scientific advisory board that reviews the soundness of the agency’s research methodology. A week later, the Interior Department announced that it would suspend the operations of 200 of its advisory panels.

The administration has mounted an attack on public statistics

Trump is certainly not the first president to remove experts who threaten his agenda. However, he has gone much further than others in trying to unleash the “anti-analytic” presidency Williams once described. In addition to de-skilling the bureaucracy, Trump has launched an unparalleled attack on the infrastructure of public statistics on which American politics plays out.

This started during the campaign, when Trump for example cast doubt on economic indicators that gauge present government performance. During the 2016 presidential campaign, Trump repeatedly said — without evidence — that unemployment figures released by the Bureau of Labor Statistics were “totally fiction.”

“Don’t believe those phony numbers when you hear 4.9 and 5 percent unemployment. The number’s probably 28, 29, as high as 35,” he told supporters.

The White House has also repeatedly cast doubt on nonpartisan forecasts of its policy proposals. In early July, and again without evidence, Trump’s budget director, Mick Mulvaney, disputed the Federal Reserve Chair Janet L. Yellen’s economic projections, suggesting that “MAGA-nomics” would yield a 3 percent annual growth rate in five years.

Nowhere was Trump’s assault on numbers more consequential than in the debate over the repeal and replacement of Obamacare. Throughout the debate over both House and Senate legislation, the White House disputed Congressional Budget Office projections suggesting that millions of Americans would lose their insurance coverage as a result of the repeal. Responding to a question about the CBO’s score, White House spokeswoman Sarah Huckabee Sanders, told reporters: “I think I know the Gospel pretty well. I would say the CBO is not the Gospel. They have been wrong before. They can certainly be wrong again.”

The administration is building an ‘adhocracy’ of doubt

Beyond merely criticizing unfavorable analyses, the administration has also begun to configure a new “adhocracy” of institutions to produce analyses that support its conclusions. The Department of Health and Human Services reportedly contracted with McKinsey & Co. to produce a heavily criticized analysis of Sen. Ted Cruz’s “health-care consumer freedom” amendment, which shows it would lower costs and raise individual market enrollment. (McKinsey has not confirmed that it was contracted to produce the analysis, citing a policy not to discuss details of client service.)

More concretely, Trump has used executive orders to create temporary organizations that will produce analyses on presidential priorities like trade, crime and deregulation. It is not uncommon for contemporary presidents to create ad hoc bodies in response to widely acknowledged public problems. At their most successful, presidential advisory commissions construct consensus about policy problems by bringing together a diverse set of stakeholders. Yet the organizations created by Trump seem unlikely to do this. Most notably, the Presidential Advisory Commission on Election Integrity is largely made up of Trump supporters and appears to exist largely to provide support for Trump’s false assertion that there was massive voter fraud in the 2016 election.

Commission reports, especially controversial ones, are more likely to end up as doorstops than policy overhauls. Shortly after releasing requests for voter information, Trump’s election commission was met with bipartisan backlash from state officials and multiple lawsuits.

Yet judging the Kobach commission by usual metrics of policy success misses the point. By further polarizing public perceptions of electoral integrity and weakening trust in vote tallies, it may change politics without changing policy at all.

This tells us something important about how Trump is changing policy. American presidents doubt or look to sideline inconvenient experts all the time. Yet Trump is going considerably further, by politicizing even very basic truth claims and looking to use his opponents’ reliance on expertise against them. It is possible that this will pay short-term political dividends, but it is likely over the longer run to have a corrosive impact on the basic structures of knowledge that government needs to do its work well.

Philip Rocco is an assistant professor of political science at Marquette University and a co-author of “Obamacare Wars: Federalism, State Politics, and the Affordable Care Act.” His current book project examines how advisory commissions reshaped the politics of U.S. federalism.

This article is one in a series supported by the MacArthur Foundation Research Network on Opening Governance that seeks to work collaboratively to increase our understanding of how to design more effective and legitimate democratic institutions using new technologies and new methods. Neither the MacArthur Foundation nor the Network is responsible for the article’s specific content. Other posts in the series can be found here.