Goodhart’s Law: Cobras, Bank Accounts & Unqualified Buyers
“Show me the incentive and I will show you the outcome.”— Charlie Munger
In a 1975 paper, British economist Charles Goodhart wrote, “Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.” This provided the genesis for Goodhart’s Law, which is commonly stated: When a measure becomes a target, it ceases to be a good measure. Anytime we use a number as a proxy for an outcome, we risk creating a host of unintended consequences. This appears to be especially true when it involves incentives.
When British Colonial officials offered a bounty to reduce the number of cobras in Dehli, it started out fine. Locals eliminated the deadly snakes faster than Rikki-Tikki-Tavi. Then something unexpected happened. Some enterprising people started breeding cobras to collect bounties. When the Brits figured this out, they ended the program. When the program ended, the cobra breeders released their vipers back into the wild. Then, déjà vu, they had a cobra problem again. Only worse than before!
A century later, the Wells Fargo banking company saw that opening new accounts correlated with profits and growth. They went all in on the metric. Branches that did well got bonuses and managers got promotions for creating accounts. Eventually motivated employees started gaming the system, opening over 1.5 million accounts without customer authorization. Can you spell “fraud”? The Consumer Financial Protection Bureau spelled it out to the tune of a $185 million fine.
The most common place we see Goodhart’s Law in real estate is when agents create incentives around a leading indicator of profit such as appointments. When inside sales agents (ISAs) receive a financial reward for setting appointments, the percentage of signed buyer and seller agreements typically goes down. Suddenly agents are saying the appointments are a waste of time, while ISAs counter that agents lack sales skills. Sorting this is like mediating which kindergartner hit who first on the playground. Savvy can agents bypass this confusion by moving the incentives to the signed agreement, or better still, the closing.
Here is the takeaway: Be wary of marrying incentives to milestones that don’t yield profit. Listings under contract, listings on the market, and listings taken tend to correlate highly to future profit. Lead generation activities may be the foundation for everything, but they are too far upstream from the closing to reward financially. In the end, you may have to triangulate lead measures to create KPIs immune to Goodhart’s Law.
For The TwentyPercenter the goal is readership. Since I can’t beam into your offices and watch you open these newsletters with glee, I have to track progress in other ways. Subscriber rates alone can be gamed. When you focus too much on the “brag number” of total subscribers, you can end up with an unengaged list, get fined for spam, or worse—blacklisted from email providers. For this newsletter, I track subscriber numbers, open rates, and unsubscribes. Together they paint a fairly accurate picture of readership.
Finally, financial incentives aren’t the only trap. Social incentives can be just as worrisome. Agents focus on public metrics like volume, units, or GCI to the detriment of profit. “Plaques over profits” is a thing. I’ve interviewed more than one top producer who destroyed their P&L pursuing production awards. Stay focused on what really matters, and make sure that your incentives reflect your priorities.
One question to ponder in your thinking time: Are any of my key metrics sabotaging my progress?