A Modest Proposal

Government agencies share all the problems they seek to fix. Why don’t they share the solutions?

Now that the election is behind us, it is time to think about disrupting the business of governing. That should get a cheer from every partisan quarter. When you start parsing the specifics of any proposed disruption, sadly, the consensus vanishes. And when you look at the etymology of disruption, it becomes obvious why. Disrupt, it turns out is not a relatively new word, born not from the Harvard Business Review, as widely believed, but more humbly with origins in medieval medicine, and before that in Sanskrit where it meant to suffer from a stomach ache.  

So what do we hate about government that we need to disrupt? There’s an old joke about two people walking down the street and as they are passing a huge government building, one turns to the other and says, “wow that’s a huge building! How many people do you think work there?” And the other says, “about 1/3rd.” Don’t worry about translating that joke for your friends and relatives who live abroad, they’ve already heard it in their native tongue. And don’t worry about offending your public servant friends and relations—it’s their joke.

And we all know what we want to disrupt in government. Perhaps the thing we are most unified on with respect to government disruption is that we want it to embrace the things we like about the private sector without the things we hate. So when we set out to disrupt the business of government, we’d like there to be more business in government, except when we don’t. That seem quite clear.

The word disruption inflicts its own stomach ache. There are, and I’m not kidding here, intense scholarly debates about the differences between disruptive innovation and sustainable innovation. The difference is typically about the initial price of the disruption and the effect it has on the overall market in that sector. Technology is usually the disrupting mechanism and Moore’s Law is the usual example. And while I am also intrigued by technological innovation (such as Estonia, the Blockchain Nation), the questions of ‘what is disruption?’ and ‘what is innovation?’ seems at best to be a distraction when applied to government.

So what do I want to disrupt (I’ll get to how very shortly)? I want to disrupt this: I want government to be about evidence-based policymaking and right now government is about policy-based evidence-making. The difference is simple. Evidence-based policymaking is about doing the things that the evidence—rigorous, transparent and objective evidence—tells you to do. Policy-based evidence-making is about knowing what you should do and finding evidence to support it—solipsistic, opaque and subjective evidence. Peter Orszag, the Budget Director early in the Obama administration noted that perhaps 1% of government expenditures were spent on evidence-base practices. If it is more today, it is probably not a lot more.

Despite the bluster from Silicon Valley, it seems to me that sustainable disruptions are typically equal part short-term changes in structure and longer-term changes in practice. A quant way of saying that would be a change in both intercept and slope. In that spirit, I’d like to propose a modest change.  

On the Importance of Sharing

Now, this topic may strike you as spectacularly naïve, and perhaps it is, but I think we’ve made some progress in the right direction and we should both continue to build on that progress and learn from it. And I don’t mean learn from it in an artificial intelligence kind of way, but in an actually intelligent way.

The thing I think we should think hard about is a piece of legislation signed into law by… President Trump. In 2019, Trump signed the Evidence-Based Policymaking Act of 2018 passed by bipartisan vote in Congress, hereafter, the Evidence Act. Now, the Evidence Act isn’t about evidence in the “research says that X has a significant effect on Y” kind of way, but rather, it is about data.

And in particular, about how federal agencies share data. Now, I should point out that the most important word in that sentence is not data, but ‘share’. Because it turns out that the federal government is terrifically, painfully bad about sharing stuff within the federal government. Exquisitely bad. Agencies in the federal government are funded separately and operated as their own fiefdom. The results can be catastrophic, such as the failure to share intelligence that might have prevented 9/11. And you may recall that the only solution anyone could think of to solve this problem was the creation of a Soviet-sized mega-bureaucracy called Homeland Security and I am certainly not advocating for more of that.

No, the Evidence Act goes in the opposite direction and is more of a way to organize a distributed system of data collection and analysis. And this is a very, very good thing. A piece of data, of course, can have multiple uses, and the idea that in 2019 we just discovered that data should be shared across agencies is disappointing, but it is also a seismic advance. The Evidence Act does all of the things a reasonable person would want it to in terms of protecting sensitive data (in a practical, non-Estonian Blockchain kind of way). But most importantly, it requires sharing.

At this point, I need to reiterate that there is almost no sharing in government. And this is really the problem that needs disrupting. When you look across the Executive Branch TO & E, you will find almost no entities that are responsible for coordinating policy. Even within the financial agencies that have oversight of many different federal agencies (OMB in particular), responsibilities tend to be agency specific. So until you go up the hierarchy all the way to the White House and the Domestic Policy Council, there really is not an executive branch entity devoted to sharing information and policy across agencies.  

The problem is that many of the biggest challenges facing the United States effect every agency. For instance, consider poverty. Obviously, poverty is a central issue for Health and Human Services. Poverty alleviation is a key part of the mission of Housing and Urban Development. And Treasury. And Justice, of course. I could go on, but you get the idea. Poverty effects Defense and Transportation too, and even the Department of the Interior.

Now poverty is just one critical issue that is cross-cutting. COVID-19, the opioid epidemic and public health generally. Education. Inequality. In fact, if you look at the four issues highlighted by the Biden-Harris transition, they all require the cooperation of multiple agencies. Climate Change. Racial Equity. Economic Recovery. Covid-19.

In my corner of the world that is concerned with reform of justice systems, we have argued forever that police and prisons cannot stop crime. Like teachers who must confront the problems their students bring to school, so to must justice systems address the problems that people in custody bring with them.  And mainly they don’t, in part because they don’t think it is their responsibility but also, other than the intelligence community, the justice system is the least likely to share data or knowledge or just about anything. There have been phenomenal efforts to create cross-agency justice coordinating bodies in the federal government, like the federal interagency Reentry Council which sought to integrate programming for people returning from prison. But this is the exception that proves the rule.

My guess is that the Biden-Harris Transition is being bombarded with groups arguing that their issue requires an interagency coordinating body. Youth. Drugs. Inequality. My guess is that if the Domestic Policy Council was to be expanded to account for all of this demand, it would have to grow an order of magnitude, if not more.

So, in the spirit of unification, let me propose a single coordinating agency that would address most if not all of the issues that need coordination across multiple agencies. That coordinating agency would address a single issue, one that is historically neglected—almost entirely—by every administration, Democratic and Republican. That coordinating agency would share data, knowledge and policy directives that would help each agency fulfill its own mission while enhancing a broad, government wide priority. It would speak to each of the issues listed above. Climate change. Racial equity. COVID-19. A more robust economy. And that issue?

Prevention.

Prevention is one of those issues that everyone is for existentially but it certainly doesn’t drive the pitchfork-wielding mob to the barricades. What is prevention? Formally, it is the removal of risk conditions and the enhancement of protective factors. We classically think of prevention in terms of vaccines, and we often use the vaccine metaphor when we talk about inoculating people against future problems. And, typically, we think about prevention as something we do for our kids.

But prevention is all of that and much more. What we have learned in the last couple of decades in particular is the role of prevention on brain chemistry. The brain, for lack of a better metaphor, is plastic, and while children’s brains are particularly malleable that plasticity persists through adulthood.  Adverse experiences—physical trauma to the brain or emotional, physical and sexual abuse can all change the chemistry of the brain in ways that lead to a myriad of bad outcomes. But other trauma changes the brain as well. Poverty, racism, drug and alcohol use and isolation can all change brain chemistry.

These problems are preventable. There is a whole science that has emerged around prevention, a large body of rigorous, objective, and transparent evidence that supports the idea that there are practices, programs and policies that have a demonstrated positive effect. Some of these focus on alleviating risk conditions and some on improving protective factors. But there is a portfolio of high performing programs and policies, just waiting for investment.

We choose instead, Democrats and Republicans, to invest in interventions. We seek to remediate problems that already exist. There are an infinity of examples of the limits of this approach, but perhaps a simple example tells the story. You have probably seen a TV or internet ad for Harvoni, which is a drug that can cure Hepatitis C. Hepatitis C is often contracted through IV drug use or risky sexual behavior, and the rates of Hepatitis C in prisons are astronomical compared to the general public. The issue is that the behaviors that are associated with contracting Hepatitis C are unaffected by the cure—those at high risk remain at high risk after the Hepatitis C virus is cleared. That there is now a cure is a tremendous advance, a godsend for those with the virus. But the limits of the intervention are clear.

Prevention is not magic. It is one path to a healthier, happier society. But the problems it seeks to prevent are intrinsic to the mission of virtually every federal agency. And prevention is expressly not an a la carte option. You cannot take a person out of an abusive situation and think their path is clear if they continue live in poverty, with exposure to environmental and social toxins. And interventions often fail to solve the recursive nature of these problems. They are like a computer program that winds its way through its source code and at the end simply starts all over. The risk conditions have to be removed and the protective factors must be sustained.

There is no natural constituency for prevention. Those who benefit from effective prevention tend not to know they have benefited. If you remove structural factors that would have committed someone to deep poverty despite their best efforts, they will rise though their best efforts and never know the challenges they avoided. Nor will all of those who contributed to their success by paying into a system that improved their environment. But we should do it anyway.

I have a lot more I could say about new structures—innovations and disruptions—that would help create a new financing system that states and local governments could use to raise capital—without raising taxes—to fund these structures. You can read a brief here, and more to come.

In the meantime, the new administration should take seriously the idea of creating an interagency department dedicated to coordinating federal prevention policy.

Why are your newsletters so infrequent?

Long, long, ago, in that instant after the Big Bang when the universe first started expanding, I worked on a series of political campaigns. I worked in the Washington, DC HQ of a failed presidential bid. To get a paid job, I first volunteered for 60 hours a week for many, many weeks, thus weaseling my way into a job with a modestly impressive title and a real salary (though far, far below minimum wage). I wore my suit to work every day and clocked astonishingly long hours, but for long stretches I had almost nothing to do. After we flamed out, I moved south to lead the field work for a Congressional campaign during their primary. We won, I left, but then was called back to be the Finance Director for the general election. One day, perhaps a month or six weeks in to the job, I had a premonition that I was about to be fired. My premonition came to me as I walked back to the office from the local bank. My wallet was empty—the bank had refused to honor my campaign paycheck and I was worried. I wasn’t worried because I had failed to fill the campaign treasury sufficiently to even pay my own salary. I was worried because I’d had no idea we were overdrawn. Later, through the kindness of my old presidential campaign colleagues, I resurfaced a month later even further south as the campaign manager of a statewide race. Failing up is good and turn out to be fairly common the campaign world.

This was an amazing opportunity for a young fella with a zest for politics, new cowboy boots, and no capacity for self-reflection. I had a quarter million dollars to spend in two months and I was completely in charge*.[1] For that campaign, we mainly did voter outreach through direct mail and radio—Cable TV was too expensive and in the days before the internet, earned media was impossible.

I got to write the radio spots.

Here’s the set-up. Our opponent was a developer who, among other things, wanted to use the decayed industrial infrastructure in the area as a foundation to build schools for the rapidly increasing number of school age children in the District. That’s a thoughtful take on an important local issue, right? Au contraire, mon frere. The radio ads went like this:

(Doom music introduction]

XXXX XXXXX is a real estate developer (sneered in a tone suggesting he might also be peddling child pornography)

XXXX XXXXX wants to teach your children in empty warehouses (at some level this might have been true)

So that builders can make a profit (Can you imagine? Businesses charging money to do work?)

[Transition to impossibly sunny music, unicorns sliding down rainbow bannisters, eating chocolate cake).

XXXX XXXXXX is the public’s watchdog (what does this even mean?)

XXXX XXXXX passed a dozen open meetings laws (content for local, open access cable!!)

With sunset provisions! (What?!? Wait. What?)

We won our race by 19 points, and ran 20 points ahead of the winner of the presidential election. A job well done. I did a few races after that, but my enthusiasm for the whole thing was diminished. I met some remarkable people, some of whom are regulars on cable news today, and I have a trunkload of amazing but true stories that I mainly don’t tell. But the whole thing was unsatisfying. It wasn’t that I had preconceived notions that were shattered—I don’t think I had any notion at all of what I was getting into. It’s just that the business of political campaigns is marketing, and I didn’t want to be a marketer. I wanted to do policy, but not policy that you put on a yard sign.

Anyway, that’s why I haven’t written much this fall.  

Musical Interlude

This historical moment definitely calls for a metal band singing Simon and Garfunkel. If you have never heard David Draiman sing, give this three minutes.

[1] *I worked for the state party and did exactly what they told me to do.