Performance Measure pitfalls

Distinguished economists, statisticians and mathematicians often have their name associated with some economic theory, concept or tool. Some examples include Giffen goods, the Nash equilibrium, the Phillips curve and the Gini coefficient. Today, I would like to introduce you to another economic concept – Goodhart’s law.

This “law” testifies that social or economic performance indicators lose their usefulness when adopted as policy targets. Put simply, when a measure becomes a target, it ceases to be a good measure. One reason for this is that people change their behaviour when they are aware of their targets. The average person is skilful enough to make targets work for them rather than against them.

Some examples include the train driver who sets off without his passengers to avoid being late, the teacher who channels her students towards easier subjects in the pursuit of better average grades or the hospital administrator who orders patients to stay in ambulances as waiting times are measured from when the patient comes through the door.

Setting performance metrics for one part of a system often leads to sub-optimal performance in the overall system. Thus, if police focus on reducing one crime measure (eg, shoplifting), other crimes increase. So, the shoplifting rate becomes a useless measure of the overall crime rate.

Goodhart’s law is also true in business. When you set up a metric by which employees are rewarded or punished, they will act to optimize that measure. The classic example here is the call centre manager who puts a time limit on customer inquiries to artificially meet his department’s time-limit target. After one question, his operators tell customers they have to phone back to get a second question answered!

Organisations use metrics for a variety of laudable purposes but often make the mistake of developing too many measures. Also, metrics should encourage behaviours that benefit the business as a whole. Enron is an example of an organisation that had (apparent) solid financial performance which masked corrupt organisational performance. Metrics should measure whether the right things are being done at the right time all the time.

Effective metrics will provide a business with the information and insight to make the right business decisions. The most powerful metrics are those that directly measure desired business outcomes. In my experience the best metrics are often subjective measures rather than numeric values. Which is why I’m inclined to hire someone who I (subjectively) believe will be a good cultural fit rather than someone who has straight A’s.

Now imagine if financial institutions around the world had metrics that encouraged behaviours which benefited the organisation as a whole. Regrettably, many Wall Street executives were remunerated largely against numeric targets (like profitability) and this monetary stimulus contributed to the destructive behaviours that led to the Global Financial Crisis.

The cliché, “Tell me how I’ll be measured and I’ll tell you how I’ll behave”, accurately describes human motivation. As I stated in a previous blog (Corporate Governance), financial institutions need to find ways to measure greed, ambition and ethics as traditional numeric targets encourage and amplify unhealthy behaviours.

There is a lesson in this for all businesses!

John (JT)Thomas

This opinion piece is provided by John (JT) Thomas, a 48-year veteran of the financial services industry and since 1987 a specialist in commercial mortgage funds. Considered by many to be the father of the modern commercial mortgage fund sector, JT helped establish and then managed – for 17 years – what became the largest and most successful commercial mortgage fund in Australia – The Howard Mortgage Trust – with assets exceeding $3 billion. Under JT’s stewardship, investors never lost one cent of their investments and indeed, investors always received competitive monthly returns. JT was also Chair of the $40 billion mortgage trust industry sector working group.

JT has been proudly involved with Princeton for nine years and sits on both the Princeton Credit Committee and the Princeton Compliance Committee as well as being an advisor to the Princeton Board.