Target Culture and The Road to Hell

Granting a person a tenured post - i.e. one from which they cannot easily be dismissed - is a big commitment for a University. So they have an easily-understandable goal (G) to appoint only people who will work actively after they have become tenured. To achieve this goal they use a proxy goal (G*) that insists that candidates must publish research papers.

A VW Golf TDI ‘clean’ diesel: the people who set emission standards hoped that they would make the world a better place; why didn’t that work?

Be honest: the VW diesel-gate scandal was not a total surprise.

And the reason it didn’t surprise you, the reason you felt you already knew it, is because it is a specific example Goodhart’s Law.

In rough terms Goodhart’s Law states that:

If you announce a specified parameter is an organisational target, then people will ‘game’ the system until that parameter becomes meaningless.

Over at ‘Less Wrong they describe the process in detail.

  • ‘Managers’ want to achieve a difficult-to-specify goal G.
  • They formulate G* which is not G, but until now in usual practice, G and G* are correlated.
  • Subordinates are given the target G* and most people try to achieve G*. As time goes on, every means of achieving G* is sought.
  • G* was formulated because it is simple and more explicit than G. Hence, the persons, processes and organizations which aim at maximising G* achieve competitive advantage over those trying to juggle both G* and G.
  • Eventually the correlation between G and G* reduces and after a point, the correlation completely breaks down.

There are many examples that I could use to illustrate the point e.g. NHS target-culture, but let’s look specifically at education and emissions to see how Goodhart’s law describes the corruption of initially noble aims.

Education

We consider that a worthy societal goal (G) is a well-educated population. But for reasons of cost and operational simplicity we substitute this with a correlated goal (G*), that students pass exams.

Initially teachers and schools try to bear in mind the overall goal (G), but after a while – and we are well into that ‘while’ – anyone attempting to achieve goal G in place of its proxy G* – passing exams – is seen as an organisational liability.

Eventually publishers get in on the act, buying up exam boards and lowering exam standards, maximising G*, but reducing progress towards the overall goal G, and minimising the correlation between G and G*.

Does that sound familiar?

Emissions

Governments have a noble goal (G) of reducing air pollution  and so demand that automobile manufacturers make cars which emit less harmful chemicals.

To be fair to all manufacturers, they specify the precise conditions under which tests are conducted and demand the results from this test (G*) meet specified standards.

The tests are designed so their results (G*) are correlated with real world performance. However the tests are necessarily simpler than reality – and can’t be overly complex otherwise the cost of measurement would be too high.

Inevitably, manufacturers make sure there cars perform well in the precise conditions of the test and tune their engines accordingly. Why would they not?

US regulations are reasonably well-devised and what VW did was (probably) illegal.

But what you have not heard is the other manufacturers stating that ‘We don’t do this’.

Why are they silent?

In general

Managers and governments face complex, multi-dimensional problems. And I sympathise. But I do not sympathise with a response to this complexity which insists on a one-dimensional focus which ‘flattens’ reality.

The people on the ground – the teachers or engineers – still face the actual complexity, but their managers deny the existence of the other dimensions of the problem. Typically they use ‘metrics’ which are supposed to be an aid to judgement, as a substitute for judgement.

This leads to workplace stress, a ‘blame’ culture and – ultimately – failure to achieve the original organisational goals.

Goodhart’s law is widely applicable, but on reflection I think it may be merely a specific implementation of the more widely-known aphorism:

The road to Hell, is paved with good intentions.

Thanks to Andrea Sella for the original tip off.

 

Advertisements

2 Responses to “Target Culture and The Road to Hell”

  1. Victor Venema Says:

    The main example in science would be the micro-management of universities and funding agencies of scientists using the number of articles published, number of citations or the Hirsch Index as a proxy for doing good science.

    Not surprisingly this has increased the number of articles published considerably. Not so sure about more scientific progress, now that scientists are forced to prioritize publishing a lot over doing what they see as most important for progress.

    We need a culture of trust.

    • protonsforbreakfast Says:

      Victor

      Yes! I had in fact written another 300 or so words on the indicators of research success which are – or are in the process of becoming – meaningless. This impacts on my daily life intensely as I am constantly forced to flatten complex topics into tickable ‘deliverables’ that are proxies for the things we are really trying to achieve. However – I try to keep the blog posts short – and I just thought I had said ‘enough’.

      Thanks again for the link about trust which is so true! And enunciated with clarity in that lecture.

      Thanks

      Michael

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: