Technocracy and the tyranny of objectivity?

By Matt Nolan 23/07/2014


First let me cover off the two reasons you have probably clicked on this post:

  1. The question mark is on purpose – even though it sounds like a statement.  In the end, these are issues of balance rather than black and white rights and wrongs.  Then again, maybe I’m biased as I see myself as a technocrat individual ;)
  2. Technocracy is an actual term for a nation governed by technocrats – I didn’t know this when I wrote it (although I did guess ;) )

I was reading twitter, as you do, when the following tweet popped up:

Objectivity in policy making, more data, rant about politics – how could I disagree!  I am an economist, I’m cynical about political parties, I attempt data analysis, and strongly support attempts at objectivity – surely our fine tweeter was talking to my soul.

And yes, data and descriptive analysis to create “knowledge” is undeniably important to the concept of informing policy making.

But I think alarm bells appear whenever politics is termed broken and objectivity is touted as a “solution”.  Especially when the critique involved appears to be pointing at someone who tends to say that we can’t just look at ways of breaking down institutions without understanding their purpose – and the ways they actually aid in coordination and welfare.  Note:  I don’t know if he said something silly today or some such, I just looked on google search and wikipedia – just as a pointer ;)

Policy conclusions are not, and cannot be, objective.  Descriptive analysis can get as close as we can manage, but we need to apply value judgments (as well as theory which may not give a unique solution in the first place faced with the data) to get a policy conclusion.  The value judgments show up when we define a “problem” – and if we aren’t clear with them we can make some awful value judgments.

This is why economics is the way it is, and why value is treated the way it is.  This is why the use of economic language of, for example, “tackling obesity” is often misplaced, and why the value judgments being used in many of these types of crusades are uncomfortable when we actually consider them.

Thinking that objective, data driven, analysis can give us our policy solutions alone is a dogmatic position that doesn’t hold up to scrutiny – it is just as bad as those who carefully think through the current institutional structure of groups but then refuse to look at data or dynamic relationships in said groups.  Or when someone spends a long time discussing why something fits in as a “problem”, but refuses to consult real world data when making the “solution”.  Again all these sides are dogmatic – and I see all of us inadvertently slip into one of these camps at times, it is a natural part of trying to understand such huge issues as a single individual!

Real policy making involves problem definition, based on a multi-dimension ethical view (Kolm, Sen) – the problem definition isn’t something we just pluck out of the air to solve, but in itself is a whole area of analysis!  Given that, we can determine what to measure, how to interpret the data, the appropriate theory, and do a “transparent” analysis that provides information – no analysis can be truly objective, but we get as close as we can when all the value judgments (assumptions) involved are clear.

Given that, the two processes of definition and analysis aid each other and it becomes important to do a third step – persuade the general public that you are providing them with knowledge, so they can decide whether this is a decision they agree with.  We are after all, a democracy.

I would note here that persuasion isn’t about hiding value judgments that make our arguments untenable to people so they vote for them – it is about giving them a weighted view of the value judgments so that they can decide whether they agree or not.  If the public doesn’t view something as a “problem” when we do, it is disappointing for the researcher – but again, we are not dictators and shouldn’t be.