Pete Michaud

Talk mathy to me, baby

Fork Pete Michaud on GitHub

Implicit Good & Externalities

March 15th, 2013

I started my first “company” when I was 13, and pretty much haven’t stopped. When I’ve taken a more traditional job, it’s been as a strategic stepping stone to being an effective and credible entrepreneur. At the same time if you had asked me when I was little what I wanted to do when I grew up, I’d have told you without blinking:

I want to save the world.

I was born to be a “social good startup guy,” and I hope to always be this naive and ambitious. A not-for-profit seems like the obvious choice, but I’ve put a great deal of thought into what a modern social good startup might look like, and the traditional models trouble me. Still, I’m left with questions.

Problems with the Traditional Model

Traditional not-for-profits operate basically like so:

  1. Acquire Money (through donations, or a for-profit arm with high margins)
  2. Spend less than 10% of that money on operations and marketing
  3. Spend the remaining money on whatever the stated mission is

A charity that exists to help homeless people might solicit donations, pay themselves modest salaries, then buy food and clothing with the remaining 90% of the money.

That’s the theory anyway.

Misaligned Incentives

The reality is that many “charities” use the vast majority of their money on “operating expenses” (read: enriching themselves). Even the many not-for-profits that try in earnest to spend as much as possible on their mission feel the pinch of the dynamic between operating effectively as an organization, and doing good.

It’s a toxic pattern in which growth is both good for the mission and terrible for the mission, and there’s ambiguity about just where the line is between overhead expense and positive returns on scale.


Even the world class organizations who successfully strike the balance between operational costs and giving, are in an unsustainable situation. Their model isn’t a machine that propels itself forward by intertia, with each part of the machine supporting and driving the other parts forward in an upward spiral. Instead it’s a one-way funnel that claws for resources it can’t economically justify having, then it gives those resources away, only to repeat the cycle. No matter how much good they do, it doesn’t help them sustain their own existence.

Emotional Bias

The model also requires that the cause pull the appropriate heart strings either of the individual donors or of the intended audience for the PR department of the institutional donor. That works sometimes, but breast cancer research is funded far out of proportion to the risk is poses and the structural problems in our world economy are complex and unsexy and they need to be solved too.

A Better Model: Implicit Good

The whole idea of capitalism is that people can trade bread and wool in a way that benefits them both. There’s an implicit good in the transaction because both parties end up warmer and fuller than they were before.

Not-for-profits exist in part because there are a lot of people who would like to trade for bread but they don’t have the wool. So someone collects the extra bread she can find from her friends and gives it to the hungry people; not for profit, but for love. But maybe next time her friends aren’t so apt to give the bread either because they don’t have it now or because they just don’t give a shit anymore.

The ideal solution, then, is to figure out a transaction that the hungry person can participate in for their own benefit and for the benefit of the other party. They used to be the recipients of charity, but once they figured out that they could sing songs in exchange for bread and wool, they no longer require fickle generosity.

They now act as their own indefinitely sustainable “charity,” engaging in implictly good transactions with other economic actors. We call that a business.

A great example of this is the idea of Fair Trade. You have impoverished people on arable land mostly in the third world who are getting pinched by lack of local demand or by fat middle men. When a large company disintermediates the market by paying directly to the farmer, the company pays less or the same, and the farmer makes more. It’s a sustainable transaction that carries an implicit good for all parties.

Easier Said

I think figuring out clever ways for previously impoverished people to participate meaningfully in the economy is the ideal, but it’s easy to say that and titanically difficult to make happen. There’s a reason the euphemism for these people is “disadvantaged”—every experience and reality of their life is working against their ability to meaningfully participate in an increasingly complex economy.

As we explore alternatives, I think it’s worth bearing in mind that this would be ideal, at those times when it turns out to be feasible.


I should also mention that there’s a different use for not-for-profits that’s not about disadvantaged individuals.

Not-for-profits also exist because sometimes, even if we collectively have the resources to trade for a good that’s collectively worthwhile, no one person is incentivized to buy it for the benefit of all. Each individual actor in the system is acting his own best interests, and the end result is a Nash Equilibrium that’s either not optimal or is actually unstable.

Not optimal” is probably a tolerable state of affairs, for example we don’t all have the buying power of billionaires even though, strictly speaking, I believe we have the material and knowledge resources for that to be hypothetically possible. “Unstable” is not tolerable, because it tends to derail everyone, for example global warming.

The trick in cases like these is that the cost that an individual must pay to for the benefit is greater than their own marginal utility for that benefit. In fact, most of the time the cost to any individual to solve the systemic problem is too high for anyone, ever, in the history of the world to solve alone.

So a company or movement has to start that either creates efficiencies in the market such that the individual cost drops below the marginal utility (think Kiva), or that aggregates resources from donors such that the cost of the high scale transaction drops below the total utility for the system (perhaps medical research).

When that’s not the possible, the idea goes, the government can warp the market (with taxes or regulations) in a way that’s advantageous to everyone overall. I could hear a rimshot in my head when I wrote that. It’s worth mentioning the possibility of a helpful government, but it’s not practical to consider on the scale of startups.

Alternative: Positive Externalities

The tragedies of the commons and disadvantaged individual have the same potential solution.

Normally when economists talk about “externalities” what they mean is a factory dumping chemicals into a river instead of disposing of them correctly. Or a majority of Walmart employees requiring government assistance to make ends meet. In both cases the true cost of doing business has been externalized to a third party, i.e. us.

What if we deliberately created positive externalities from transactions that already benefit the parties involved?

We’d have two actors trading for mutual benefit (sustainable), with another actor incidentally benefitting, all of them enjoying the implicit good of the transaction.

It’d be like if the factory found someone who really needed those chemicals, and would take them for free. Everyone would win.

One great example of a positive externality that cool tech kids would understand is that when a given user transacts with a social company like Instagram, the value for all users goes up. Each user benefits from the whole network of users being larger, hence the term “network effect.”

The problem is that the external benefit in this case is shared only among customers.

Here are my questions:

  1. How can we arrange other industries to externalize value for people who need that value and can’t otherwise get it?
  2. Can we figure out a way to create and capture that external value systematically, across many industries, or is it a matter of discovering idiosyncratic industries with incidentally useful waste products?
  3. If a system could be devised, would the system itself be able to capture value such that it could be a self sustaining? In other words, could you create a profitable company that redirects other companys’ externalities to be of use to underserved populations?

I’m trying to figure out how to build the system in #2 so I can determine the answer to #3. If the system in #2 is possible, then it’ll be incredibly powerful even if it relies on the traditional not-for-profit model. If #3 is possible as well then it could be a game changing juggernaut of social good. I want in on that.