Opposite is not a winning strategy
Dems need alternatives whose strength comes from a positive vision of what to do, not defined by what not to do.
You should never let your enemy define you, but it’s also a mistake to define yourself solely in relation to your enemy, especially as the opposite. I see this in some of the opposition to Trump, a knee-jerk reaction that “if he is for X, then we are against X.” This happened during the pandemic, and led many on the left to take positions that proved unwise. Monica Gandhi, a professor of medicine at UC San Francisco, called this out in the Atlantic back in 2021. “If he said, ‘Keep schools open,’ then, well, we’re going to do everything in our power to keep schools closed,” she told Emma Green. It was not good that blue states kept schools closed for so long — not good for kids, not good for parents, and not good for the reputation of Democrats nationally. Gandhi describes herself as “left of left,” but was heavily criticized by the left for her advice to re-open. It was hard for others on the left to see her position as a considered public health stance that had weighed pros and cons. The fact that she happened to agree with Trump eclipsed all else.
“We have to be as different from him as possible,” a friend of mine told her children when Trump was reelected. When faced with cruelty and rashness, of course it makes sense to double down on kindness and thoughtfulness. But reflexive reaction to an extreme has not served Democrats well. Because Trump had governed carelessly, recklessly, and unilaterally in his first term, the Biden administration was determined to act with enormous care, caution, and consensus. Jake Sullivan recently described this dynamic to Ezra Klein. When Ezra asked him “where is all this resistance to speed coming from in a government you supposedly control?” Sullivan responded:
“It takes a couple dozen people to say yes to make something happen, and it only takes one person to say no to stop that thing from happening. The bias is always toward no. And you might ask: Why can’t the president just override the no? That’s where we as an administration were intensely scrupulous about process, propriety, mindful of the role of the agencies, and so there was a degree of self-deterrence that was almost culturally built in.”
Scrupulousness about process can be a good thing, in moderation, and in the right circumstances. But too much care for process and procedure led to things like the Justice Department’s slow and ultimately failed pursuit of the January 6th case, as Quinta Jurecic’s brilliant piece in the Times points out. It also led to a lack of promised EV chargers, a lack of rural broadband connections, and a lack of new green energy projects permitted, as Ezra and his co-author Derek Thompson have pointed out on countless podcasts over the past week as they promoted Abundance. This “self-deterrence” Sullivan describes is part of why Trump got reelected. Being “as different as possible” – being in opposite mode — kept Democrats from bringing key policy priorities to fruition in ways that voters could see and feel.
Opposite mode isn’t the right reaction to DOGE either, though a response is certainly needed. It seems that concerns about the DOGE team’s handling of data have been well founded. Last week, a judge ruled that DOGE violated the Privacy Act in its handling of data at the Social Security Administration. Thoughtful opposition to this abuse should involve the enforcement of our laws and consequences for those who break them. But well-meaning efforts led by Democrats in the House and Senate appear poised to try to lock down data-sharing more generally. And while Dem-led bills are unlikely to move in this Congress, today’s discussions and compromises lay the groundwork for future legislation. Making data sharing a boogeyman today will taint future debates about automatic benefits qualification, “ask once” data approaches, and the ability to model policy approaches across silos to improve outcomes.
Before DOGE, most people in my government service delivery circles felt that we needed better ways to share data between and among agencies. You can really improve the experience of interacting with the federal government if you use administrative data wisely. One example is understanding who might be eligible for what benefits. Another is not asking information from an individual or business if a different agency already has it. Before DOGE, we all wanted an IRS API (an Application Programming Interface, which I’ll define more in a second), because of all the ways it could have made government services easier to use and administer. But now that DOGE is planning one, the left has decided it’s dangerous —- not just the way DOGE wants to do it, but the whole idea. Makena Kelly writes of the proposed IRS plans in Wired:
A “mega API” could potentially allow someone with access to export all IRS data to the systems of their choosing, including private entities. If that person also had access to other interoperable datasets at separate government agencies, they could compare them against IRS data for their own purposes.
But not being able to export all the data in a system is why APIs exist. APIs are designed to provide specific, limited functionality through predefined endpoints with proper authorization controls. They're meant to facilitate secure, controlled data exchange between systems. Properly designed APIs include: authentication (verifying identity), authorization (permission levels), rate limiting (preventing excessive requests), data filtering (returning only necessary information), and audit logging (tracking all access). A core principle of API design is to return only the data needed for a specific task, not entire databases. (For example, an agency might need to know if a company filed taxes in the previous year. An API might return a simple yes or no without divulging other sensitive information.) Critical systems are often segmented, requiring multiple levels of authentication and specific permissions to access different data types. It is possible, I think, to design an API that does none of these things and just lets whoever wants to export all the data in a system, but I’m not sure at that point you’d even call it an API. Wired editors of all people know this. They would have cheered an API under democratic leadership, but in opposing abuses of data they’ve implicated APIs generally, effectively throwing the baby out with the bathwater. Hold DOGE accountable to appropriate use of data, absolutely, but don’t let opposite mode make it harder for a future administration to build a responsible API.
If DOGE is violating privacy and security laws, new stricter rules are unlikely to constrain them. If they are breaking rules and it’s being tolerated, it may be in part because those rules were so strict to begin with — or in this case, so unclear and on a practical basis interpreted overly strictly, given that the Privacy Act was written in 1974, when data sharing meant something entirely different— that they seemed to be working against popular outcomes and lost support. But new stricter rules will absolutely constrain a future Democratic or other rule-following administration. We should write the laws we think we need to govern well regardless of who is in power. If Democrats feel that laws are being broken and Republicans are ignoring these transgressions, their job is to win enough seats in the mid-terms that they can do something about that. Attempting to strengthen laws that a rogue administration can ignore anyway may satisfy a resistance-focused base, but it’s neither a short-term swing vote winning strategy nor a long-term governing strategy.
In some cases, Dems seem to have drifted into an opposite stance through inaction. It was clear since at least November (arguably before), that there would be huge cuts to the federal workforce. Dems could have developed a strategy, one that might have involved looking across federal agencies, identifying the less critical programs, and proposing to protect the most valuable activities and staff by directing the cuts where they’d have the least negative impact. The Trump team may have entirely ignored Dem proposals, but the American people should have at least heard that one party had a vision for responsible disruption. Instead, Dems’ objections to DOGE’s destruction read as an across the board defense of a sprawling bureaucracy, not all of which is worth defending.
I was recently digging around in reginfo.gov, as one does, trying to get my head around how OIRA administers the Paperwork Reduction Act. I was just trying to understand the kinds of documentation that an agency has to compile to get permission from OIRA before launching an information collection, which is usually some sort of web form, so it didn’t really matter what program I was looking at. Somehow I ended up down the rabbit hole of a website called the USDA Recipe Finder. I was exploring the clearance documents for two forms: one to allow users to submit recipes and another to allow them to review them, when I realized I should look at the site in question for some context.
The USDA Recipe Finder is part of a program called SNAP-Ed, which may be a very valuable program overall — I don’t know. But its website is not great. On the homepage I am told “Education can be used to deliver nutrition messages to the SNAP-Ed audience. Approaches include Policy, Systems, and Environmental Change and Social Marketing. See how SNAP-Ed programs have used Nutrition Education to communicate healthy messages to SNAP-Ed audiences.” I am then given three options to click: SNAP-Ed Recipes, Healthy, Thrifty Holiday Menus, SNAP Recipes at MyPlate Kitchen. What’s the difference? Other than one being holiday themed, it’s unclear.
Clicking through doesn’t solve the mystery. I choose the first option and am offered an oddly similar menu: Recipes from SNAP-Ed Partners, Recipe Video Collections, SNAP Recipes at MyPlate Kitchen, Healthy, Thrifty Holiday Menu Ideas, Most Popular Spanish Recipes, and Seasonal Recipes. I click on the first option again, and see the logos of 44 state SNAP-Ed partners, with some states missing and some states represented twice. Why should I choose a recipe by which state cataloged it? Click on any of these and you’ll finally find yourself in a recipe finder for that state or that program within that state. Being a former Californian, I clicked on CalFresh, and then the first category on the CalFresh HealthyLiving recipe finder, for “drinks and smoothies.” The first recipe there was for “Cucumber Mint Breeze,” which I offer to you here in full.
Cucumber Mint Breeze
INGREDIENTS
½ cup sliced cucumbers
1-2 sprigs of fresh mint
Ice
Preparation
Fill pitcher halfway with ice.
Add sliced cucumbers and mint.
Fill with water. Chill for at least 20 minutes before serving.
Store in refrigerator and drink within 24 hours.
Makes 4 servings - 1 cup per serving.
Delicious? Yes? A good use of taxpayer dollars? Not in a world where recipes, even healthy ones, are ridiculously easy to find. And advice to put mint and cucumber into water if you want mint- and cucumber-flavored water reminds me of the UK’s Government Digital Service taking down a page that counseled Brits that when it was cold out, they should consider wearing a jumper (that’s British for sweater.) The creation of a good gov.uk required the removal of a lot of bad content. If you’ve heard me say we need to subtract more than we add, this is just part of what I mean.
To be fair, this particular recipe is on a site funded by the State of California, not the federal government, unless California used federal funds to make it. The recipe finder is just how we got there. But that almost makes it worse, because it reminds us of how many different taxpayer-funded recipe finder sites there seem to be across state government (at least 44) and federal government (at least two: SNAP-Ed and MyPlate). No, wait. I just typed “healthy recipes” into google and found, among many, many professional and user-generated websites dedicated to healthy cooking, that there’s also nutrition.gov, also from USDA. So that’s at least three. There’s a lot of referring back and forth, but also a lot of content across these many sites.
There are many valid reasons why we might want to educate SNAP users about healthy eating choices. But I’ll go out on a limb here and say our government could function without the USDA Recipe Finder. I don’t think it should function without SNAP itself, and while I am certain the money we spend on either the USDA Recipe Finder specifically or even SNAP-Ed more broadly is tiny compared to what we spend on SNAP payments to eligible recipients, if staff cuts are going to happen, it would make a lot of sense to direct those cuts away from those who administer a life-saving program and towards things we could live without. Looming cuts could have been an opportunity for Dems to articulate what matters most and commit to strategic protection.
Absent that message, Democrats run the risk of standing for no cuts, for keeping it all, regardless of the cost or the return on investment — essentially, for the status quo. DOGE is not popular, signalling that many voters want more responsible change, but they still want change. Recall that in the run-up to the election, voters strongly preferred anything but the status quo. When asked “which is more important now, preserving America’s institutions or delivering change that improves Americans’ lives?” 78% chose change. We can’t ask voters, especially the voters Democrats need in the midterms, to again choose between what we now know is scary change and the perception of no change. Democrats must offer an alternative, and particularly a positive, alternative vision of government reform. Being in opposite mode deprives Dems of the chance to shape the conversation and sends their favorability ratings through the floor.

None of this is an argument for Democrats to be more like Trump, or Republicans, or anyone else. It’s an argument for having an independent viewpoint informed by a consistent and examined set of values. It’s an argument against knee-jerk responses. It’s an argument against defending a status quo that simply isn’t worth defending. Most of all, it’s an argument for being a party that wins elections because it governs wisely, and gets to govern wisely because it wins elections. Dems can’t be that party if they’re stuck in opposite mode.
If Democrats want to tackle government reform, my colleague and I offer a roadmap for that. More broadly, the so-called “abundance agenda” offers fresh thinking that promises to address real problems that voters feel viscerally. A bold, ambitious, but responsible government reform agenda is different from what Trump is doing with DOGE, and abundance-oriented policies are different from the scarcity that Trump’s economic policies are bringing, but both appeal to a wide range of ideologies and have the real potential for bipartisan support. Both are alternatives whose strength comes from a positive vision of what to do, not defined by what not to do.
As the new series The Comeback reminds us, opposite mode is what kept the Red Sox losing against the Yankees. When they woke up every day trying to beat the Yankees, they kept losing. When they let go of that, and just tried to be the best team they could be, that’s when they won their first World Series in 86 years. Democrats won’t win being the opposite of anything; they just need to be the best team they can be.
Great post and I also enjoyed your conversation with Tyler Cowen. Where I get stuck is “what can I, a frustrated voter with a strong interest in policy who wants the government to *actually work,* do? Now? The sorry state of the Democratic party is alarming.
I appreciate your focus on outcomes for the American people and not simply on winning elections. The goal should be to come up with ideas that allow for government to create opportunity for the American people. I am looking forward to reading your full report. Just glancing at its contents it illuminates many of my observations from my 18 year career in Defense Policy. DoD is often turned to as a solution to state problems because it has state capacity. When DoD has a need to accomplish a new mission it has been able to get resources to create the capacity it needs. Capacity is critical to effectiveness.