I joined the Economic Research Service in 2016. I wanted to use my academic training to do something in the public interest, but I didn’t really expect to get involved in agriculture. Then I got absorbed in the subject: Humanity’s dependence on the environment is made explicit through our food systems; without the right combination of weather, soil and labor, nobody eats.

Most people don’t need to think frequently, or ever, about the economics of honeybee pollination routes or the cost of the Federal Crop Insurance Program. But if they eat almonds (which are pollinated by bees) or pay taxes (which subsidize farm insurance), they need experts to make sure that food systems work efficiently and public funds are spent effectively. At ERS, we studied all aspects of food production, occupying an obscure but important niche: Many of our research topics wouldn’t make for an exciting academic tenure file but had huge implications for policy.

Out of the blue, in August 2018, Agriculture Secretary Sonny Perdue announced that my agency and the National Institute of Food and Agriculture would relocate from Washington to some yet-to-be-determined location. He claimed this would lower costs and bring us closer to “stakeholders.” That stated justification was a fig leaf for the administration’s true intentions. We didn’t need to sit next to a cornfield to analyze agricultural policy, and Perdue knew that. He wanted researchers to quit their jobs.

The administration had demonstrated its hostility to our agency. Their proposed 2020 budget would halve our staff from 329 to 160, slashing “low priority research” areas like food assistance programs and conservation efforts. The department had started requiring us to add disclaimers to our scientific journal publications, even those that passed peer review, undermining the authority of our own work.

I didn’t take the threat seriously at first: Federal relocations tend to be incredibly complicated, and I basically believed that the Trump administration’s incompetence would get in the way of its malevolence. But apparently, the political appointees in the Office of the Secretary were motivated. They didn’t like it when our research was at odds with the administration’s narratives. (The Agriculture Department wanted to restrict access to food stamps, for example. According to our models, however, food assistance programs were often a positive multiplier for local economies.) They certainly didn’t appreciate my colleagues’ paper showing that the 2017 tax cut would give the biggest benefits to the wealthiest farmers.

Rather than try to quash individual studies, the administration decided to try to drive out the researchers wholesale. Perdue insinuated that our research was politically motivated, telling reporters that the department needed to avoid past mistakes in which decisions were made “based on political science rather than on sound science.” By suppressing scientists instead of their science, he could avoid engaging with the specifics of our work.

My colleagues and I organized letter-writing campaigns, contacted our congressional representatives and voted to unionize. I tried to see the bright side: I’d always liked my job, but if the administration was going to these lengths to try to silence us, maybe our work was even more important than we thought. Inspired by the big floods in the Midwest, I started working on a machine learning tool that could ingest large amounts of data and use it to predict when and what farmers plant. Eventually, it could be the beginning of a larger framework for integrating farmer behavior and economics into climate change research. I acted as though I would be there for the long haul, even as the desks around me were emptying.

Then, in June, Perdue announced that the agencies would move to Kansas City, Mo. I went on the job market.

My calendar filled up quickly: I had 42 interviews in two months. Eventually, I accepted a position as a data scientist at the University of Pennsylvania’s medical system. It’s a great job and an opportunity to move back to Philadelphia, where I was born and raised and where my parents still live. Predictive medicine, my new field, is fascinating and important. At ERS, though, I was one of the very few people in the government who knew how to apply machine learning tools to problems of agriculture and climate change, and I had colleagues who had honed even more arcane backgrounds and skill sets. Now, the public will lose the targeted expertise we developed on their behalf.

Most of my colleagues have moved on. Contrary to the common talking points about cushy government jobs, we all knew that we could have gotten higher salaries in the private sector, faster advancement elsewhere in the government or more perks in academia. We loved ERS because it offered a rare degree of intellectual freedom, combined with the chance to make a real impact. We got to spend a great deal of our time pursuing research questions that we defined. The rest of the time, instead of logging service hours by sitting on committees, teaching or grading like we would at a university, we got to advise on policies that affected people’s lives.

That’s over. Our union estimates that of 180 employees who were assigned to relocate, 141 declined. They weren’t willing to uproot their families, sacrifice their spouses’ careers, or in some cases, disrupt their medical treatment to stay at an agency that remained firmly in Trump’s crosshairs. The agency has managed to hire a few sharp new researchers in Kansas City, but they’re just a drop in the bucket compared to what’s being lost, and it’ll take them a long time to learn their fields.

All of the people who study genetically modified organisms left. The team that studies patent law and innovation is gone. Experts on trade and international development, farm finance and taxes all left. Many people transferred to other agencies in the USDA, where they’ll help implement programs but will no longer have a mandate to produce the essential research that’s needed for sound policymaking. The publishing staff all left, delaying dozens of reports on subjects from veterans’ diets to organic foods. Projects that have been years in the making, on issues from honeybees to potentially harmful herbicides, will never see the light of day.

Around two-thirds of ERS positions are vacant. With all those seats unfilled, the handful of remaining staff will spend their days buried under policy review requests, too busy putting out fires to work on challenging new questions. When I joined in 2016, the agency already faced an aging workforce and was poised to rapidly recruit more fresh PhDs. Now, many of its veterans are retiring early, before they can pass on their knowledge. That depletion will set the agency back by a decade or more.

The agency never has a perfectly smooth relationship with any White House: Its studies have contradicted rationales for policy ideas ranging from biofuels to farm subsidies. But the Trump administration seems singularly and openly opposed to its basic existence. They can’t tolerate it when scientists present hard truths they don’t like. If lawmakers want to know about, say, the effects of tariffs on the broiler chicken industry or the impact of farm conservation payments on the dead zone in the Gulf of Mexico — obscure topics that can mean millions of dollars and thousands of jobs — they’ll be operating in the dark.

As told to Post editor Sophia Nguyen.