When Economic Metrics Stop Telling the Truth: Goodhart’s Law
Author: Inesh Tickoo
Policymakers and firms rely on indicators to guide decisions. Goodhart’s Law identifies a
limitation of this approach, that when measures start being targets, they stop being good measures.
Let me explain that more clearly.
A speedometer tells you how fast you’re going. A speed limit changes how you drive. Once a speed
limit exists, it changes drivers’ behavior to avoid tickets, like slowing down when they see the cops.
When a statistic is used to observe the world and its reality, it behaves like a speedometer. When the
same statistic is used to judge performance or trigger rewards and punishments, it’s like a speed limit.
Inflation statistics like the Consumer Price Index (CPI) track the cost of living. When the Fed
says “we want 2.1% inflation,” businesses start setting prices and wages with that number in mind.
This statistic now affects strategy and sets expectations rather than indicating underlying conditions.
Goodhart’s Law also applies to labor markets and productivity. Firms that focus on per-hour
output or quarterly efficiency targets may boost short-term productivity through moves like cutting
training, delaying maintenance, or increasing employee workloads. Cutting training might save time
and money this quarter, but workers may turn out to be less skilled later. Delaying maintenance
will keep machines running today but they might break down tomorrow. Increasing workloads
might squeeze more output out of workers for now, but it leads to burnout, mistakes, or turnover.
While a few metrics improve, the underlying health of the firm or economy may deteriorate, leading
to weaker long-term growth. At the same time, expansions can sometimes mask fragility. Strong
headline indicators may encourage risk-taking and leverage, even if deeper structural problems are
building beneath the surface. When conditions shift, those hidden weaknesses become visible, often
abruptly.
Overall, Goodhart’s Law says it’s important to use economic indicators for diagnostics rather
than goals. Metrics and statistics are essential for understanding the economy, but overreliance on
any single measure can distort incentives and reduce the reliability of the signals policymakers and
investors depend on.