When I studied economics, one of the concepts that struck me the most was the concept of externalities. This International Monetary Fund post explains it well. In short, externalities are costs or benefits of an economic activity that affect third parties who did not choose to incur them, leading to a divergence between private and social costs or benefits. They’re spillover effects—positive or negative—that the market price fails to reflect. A classic example is air pollution from a factory, where nearby residents bear health and environmental costs not included in the price of the factory’s products.

Open source is full of externalities. On the positive side, adoption creates ecosystems of developers and provides many paths of distribution. On the negative side, there’s often underinvestment in the very projects that sustain the ecosystem. I have a lot of empathy for why, when open source meets finance and private equity, things can go sideways. You can look at a business built on open source and see seemingly amazing margins—efficient R&D that compounds in a DCF model. A percent here or there over many years really adds up.

My plea to investors in open-source businesses is this: when a business is built on top of open source, incorporate a restorative investment percentage back into the projects critical to the end-user experience of what you’re offering customers. In WordPress, we call this Five for the Future, but it doesn’t have to be five percent; it could be 0.1%. Plan for it when modeling your expected IRR hurdle from an investment. Then, a few years down the line, when the small percentages start to add up, you won’t face a big catch-up or gap.

This underinvestment is itself an externality. It doesn’t appear on the balance sheet, but it can manifest in black swan events, such as security breaches or remote code exploits. Technical debt is one of the largest unaccounted-for externalities in the world today. Engineering, in the long run, is primarily a craft of maintenance rather than creation. The bulk of the cost of something comes from its upkeep over time.