Count Who Stays: Retention as the Real Diversity Metric in Environmental Work

In 2007, Felicia Marcus, then the chief operating officer of the Trust for Public Land, wrote an essay for a Yale collection on diversity and the future of the US environmental movement. One of her observations still lands: conservation groups had made sincere efforts to hire people of color, and too often those hires did not stay long. Our explainer on her essay covers the rest of her argument. That single point, made almost two decades ago, is the place to start any honest conversation about diversity in environmental work today.

Hiring is visible. It produces announcements, new headshots on the staff page and numbers that look good in an annual report. Retention is quieter. It shows up as a farewell lunch, a vacant position, a gap in institutional memory. An organization that hires five people of color in a year and loses four has done something, but it has not changed who it is.

What the 2014 numbers showed

The study most people cite is The State of Diversity in Environmental Organizations, published in 2014 by Dorceta Taylor, then at the University of Michigan, for Green 2.0, a campaign pressing the sector to diversify. Taylor's team gathered data from almost three hundred mainstream environmental nonprofits, government environmental agencies and grantmaking foundations. Across those institutions, people of color held no more than about 16 percent of staff and board positions, at a time when they made up more than a third of the US population. Taylor described the barrier as a green ceiling.

Several of the study's other findings point straight at retention. Many organizations recruited through word of mouth and existing networks, which tends to reproduce the staff they already have. Gains in gender diversity had gone mostly to white women. And while most organizations said they valued diversity, relatively few had dedicated staff or concrete plans to act on it. Our guide to landmark reports summarizes the study in more detail.

Since then, Green 2.0 has published regular report cards on the demographics of large environmental nonprofits and foundations, which are available on its website. Those reports, and a decade of sector conversation, have pushed many organizations to widen recruiting. Fewer have looked hard at what happens after the offer letter.

Why hiring alone did not hold

The usual explanation for slow progress was the pipeline: not enough qualified candidates of color. Taylor's findings on insular recruiting undercut that from the start, and the pipeline story has a second weakness. It says nothing about why people who were hired, qualified and motivated, later left.

The pattern will sound familiar to many people in the field. Someone is hired into a junior or program role, often in community engagement or outreach. They are asked, formally or not, to carry the organization's diversity work on top of the job. Their pay is set without much transparency. The path upward is unclear or runs through informal relationships they are not part of. After two or three years they take their skills to a city agency, a foundation, a community group or another sector entirely.

None of that is unique to environmental work. But the sector's structure makes it worse. Many organizations are small, with flat hierarchies and few senior roles. Program staff often work in the field, far from where decisions are made. And a culture built around a particular idea of the outdoors can make people who don't share that background feel like guests.

That matters even more in 2025. With federal agencies dismantling diversity programs and some funders and nonprofits growing cautious about the language of DEI, formal initiatives in many organizations are shrinking or being renamed. The practices below are mostly plain good management. They hold up whatever the political weather, and they work best when no one has to defend them as a special program.

Being the only one in the room

In her 1977 study Men and Women of the Corporation, the sociologist Rosabeth Moss Kanter described what happens to people she called tokens: members of a group that makes up only a small fraction of an organization, roughly 15 percent or less. Tokens are highly visible, so their mistakes are noticed and remembered. Differences between them and the majority get exaggerated. And they are pushed toward stereotyped roles. It is striking how closely Kanter's threshold matches the ceiling Taylor found in environmental organizations decades later.

A related idea comes from the psychologist Amado Padilla, who in 1994 described cultural taxation: the extra, often unpaid, service that faculty of color were expected to provide simply because of who they were. Environmental staff know the pattern well. The only Latino staffer translates the flyer. The only Black staffer is invited to every diversity panel. The work is real and valuable, and it rarely shows up in a job description, a performance review or a raise.

What helps:

• Hire in clusters where you can. Two or three people of color joining a team at once changes the dynamics far more than one at a time.
• Pay for extra service. If someone leads an employee group, sits on a hiring committee for equity reasons or translates materials, count it as part of the job, with time and compensation to match.
• Fund affinity groups properly, with a small budget, time during work hours and a senior leader who listens to what they raise.
• Don't make one person the spokesperson. Rotate external speaking roles, and ask before volunteering anyone.

Pay: publish the range and audit the result

Pay is where good intentions often fall apart. Environmental nonprofits run on tight budgets and tend to set salaries case by case, which rewards people who negotiate hard and people who can afford to accept low pay at the start of a career. Neither group is distributed evenly by race or class.

A growing number of states, including Colorado, California, Washington and New York, now require employers to include pay ranges in job postings. Even where the law doesn't require it, the practice is worth adopting everywhere. Publish salary bands for every level. Stop asking about salary history. Then, once a year, compare median pay by race and gender within the same level and role, and close any gaps you find. Write down how raises are decided so the answer doesn't depend on who asks.

Mentors give advice; sponsors spend capital

Sylvia Ann Hewlett's 2013 book Forget a Mentor, Find a Sponsor popularized a distinction that every organization should borrow. A mentor offers advice and support. A sponsor uses their own standing to advance someone: putting their name forward for a promotion, handing them a high-profile project, defending them in meetings they are not in.

Many staff of color in environmental organizations have mentors, often other people of color elsewhere in the sector. Far fewer have senior sponsors inside their own organization. To change that, make sponsorship a leadership expectation rather than a personal favor. Ask each senior leader whom they are actively advancing, and track who gets the assignments that build careers: presenting to the board, managing a budget, leading a major partnership, speaking for the organization in public.

Build a ladder people can see

In a 20-person organization, the only way up is often out. That doesn't have to be true. Define levels for each job family, with titles, pay bands and the skills expected at each step. Post internal openings before external ones. Offer acting roles when a manager is on leave. Set aside a development budget for each person and make sure it gets spent. For field staff, create routes into planning, policy and management that don't require leaving the work they love.

Clear ladders help everyone, but they help most the people who were never part of the informal conversations where opportunities used to be handed out.

Exit interviews come too late; stay interviews don't

Exit interviews have two problems. People leaving tend to soften their answers to protect references and relationships, and the results often sit in a file. If you use them, have someone outside the reporting line conduct them, follow up a few months after departure when people speak more freely, and report the themes to leadership and the board.

Stay interviews do the same job earlier. Once or twice a year, a manager sits down with each staff member and asks a few direct questions:

1. What do you look forward to most in your work here?
2. What might make you think about leaving?
3. When did you last feel your work was recognized, and by whom?
4. Are you doing work that isn't reflected in your role or your pay?
5. What assignment or skill do you want next, and who here could help you get it?
6. Can you disagree with your manager or leadership without paying a price?
7. What one change would make your job better?

The conversation only works if something happens afterward. A stay interview followed by no change teaches people that speaking up is pointless, which is worse than not asking.

What to track

These measures work for most organizations. Split each by race and ethnicity, gender and job level where numbers allow.

Small organizations, small numbers

Most environmental groups are small, and that creates a privacy problem. In a 15-person organization, reporting retention by race can identify individuals. Combine years, report ranges, share detailed figures only with a small group bound by confidentiality, and lean on stay-interview themes when the numbers are too thin to mean much. Being careful with data is part of earning trust, which is the point of the whole exercise.

If you can do only one thing this year, hold stay interviews with every staff member, then tell the whole staff what you heard and what will change. For further reading on building more inclusive organizations, see our resources page.

Next: browse the full Pack list.