Monthly Commentaries

Goodhart’s Law

Download PDF

Our February comment was released following a difficult final ten days of the month – not that it relates to the current events. The events are, as it’s said, what they are. The market already understands that virtually everything in China is shut down. Autos sales have plummeted, 99% of nightclubs closed, 70% of restaurants closed, 50% of Mom & Pop stores are closed. The list goes on. What investors are now starting to consider is the second derivative – where the opportunities lie.

“Believe me, there are eternal investment principles, and technology doesn’t alter them: Time is your friend. Impulse is your enemy. Buy right and hold tight. Cost matters. If you aren’t sure, diversify. Invest for the long-term. Stay the course. “ –John C. Bogle, Reinventing Mutual Funds, speech, June 11, 2001,

“Figures often beguile me, particularly when I have the arranging of them myself…”

Mark Twain

Every quarter, economists await the Chinese government’s regularly‐scheduled release of GDP data, to glean insights into the world’s fastest‐growing major economy. Conventionally, GDP is the best proxy for how well a national economy is faring. However, with China there is a slight wrinkle: it is the only country in the world that formally has a specific level of GDP growth as a target—and this targeting creates biases. Statistical inflation begins at the local level and continues upwards.

When data is biased to be systematically incorrect, errors compound. The Brookings Institution estimates China’s growth has been overstated by about 2 percentage points annually in recent years, and overall its economy is about 12 per cent smaller than indicated by official figures.

In 2015, several Northeastern provinces of China admitted faking GDP growth figures, and later in 2017, two more provinces were found faking economic data. Inner Mongolia and Liaoning are among the provinces that most overestimate their local GDP, inflated by 20% and 17% respectively. China’s data is based on information provided by local governments, whose officials are rewarded for the over‐achievement of growth targets.

Because of the skepticism towards official Chinese GDP data, economists have developed other measures. For example, the “Li Index” tracked by The Economist magazine has three inputs: electricity production, rail cargo shipments, and bank loan disbursements. It’s named for Li Keqiang, current Premier of China, who told the American ambassador to China that these particular indicators were more reliable than official economic data.

The data quality problems created by China’s GDP targeting is an illustration of Goodhart’s Law: “When a measure becomes a target, it ceases to be a good measure.” Actors aware of a system of rewards and punishments optimize their actions to maximize their own rewards.

Mundane examples of Goodhart’s Law pop up in many workplaces. Consider a call centre that sets a target for the number of calls employees must complete per hour. Yes, this motivates the employees to complete as many calls as possible, but they will now be more concerned with getting the customer off the phone than they are with solving his or her problem! Participants exploit opportunities to game the system—and the system designer’s goals are thwarted.

Goodhart’s Law often comes into play when public companies set internal targets. Whenever we see a stated target, we reflect: does this target actually produces an outcome that is economically meaningful to us as shareholders?

Some years ago we met with the management team of a Canadian public company. In the meeting, there was a stated three‐year target to generate $2 billion in annual revenue, five times their existing run rate. Exciting, until you ask a few questions. Like how many dilutive shares would they have to issue to finance the growth? How much debt would they have to layer on the balance sheet? And, is the new revenue coming at the same margin as existing revenue? Revenue growth is typically a sign of business health, but should be an outcome of pursuing profitable growth, not viewed as an end in and of itself.

As with most things in corporate life, the leader sets the tone, which is why in our investment process we spend a lot of time examining the CEO’s track record of setting up the right goalposts for employees to kick the ball through. As Sam Altman, president of startup incubator Y Combinator, said in a lecture to Stanford entrepreneurship students: “It really is true that the company will build whatever the CEO decides to measure.”