This is my final article on interest.co.nz.  Although the rest of Infometrics will be continuing to pop up articles for everyone to read, I’d like to thank you all for reading and commenting – it has been much appreciated and good fun.

Recently, I have seen more and more writers rail against gross domestic product (GDP), saying that we need clearer indicators and output targets to set policy with.  On one level this is true, GDP is overused and abused.  However, these writers are really falling into the same trap as those who excessively obsess themselves with GDP statistics – they are overly focused on specific outcomes, and are ignoring the trade-offs associated with their choices. 

At first brush, the comments in the previous paragraph may seem incredible.  I work for an economics consultancy, and spend a large chunk of my day writing about GDP.  Virtually everyone outside of the economics discipline, be they from the left or the right, believes that economists have an unhealthy obsession with GDP statistics and thereby tend to “know the price of everything, but the value of nothing”.  This can be seen as targeting the GDP ‘machines’.

Given this view of economists, the solution seems simple enough – take the GDP creating machine view of the economy, and include other things that really represent value.  If we can stick these all together into a ‘happiness index’, then policy makers can come out and ensure maximum happiness for all!  Think of this as a ‘happiness machine’.

This sounds progressive, forward looking, futuristic, powerful, and constructive.  It binds together the corporate organisational sentiment in the New Zealand right, with the state driven constructivism of the New Zealand left.  Outputs, targets, and the happiness machine – this is a way forward for New Zealand policy. 

While such projects create work for economists and sociologists, there are two fundamental principles that economists use that are being understated:  trade-offs and the fundamental limits to knowledge.  And understating these limits when determining policy is a sure fire way to ensure we inadvertently hurt people.

What are these machines?

Let us start with the example of how people treat plain GDP at the moment.  In many people’s eyes, GDP is the economy – we are a GDP machine, creating bits of GDP that we consume or use to make other future bits of GDP.  In this world people say “let us do things that will make more GDP”, politicians promise to do things that will make more GDP, interest groups sell their ideas on making more GDP.

Many individuals do see past this, and in turn accuse economists of an unhealthy obsession with GDP – but in truth it is the way GDP is reported and treated that is excessive, not the non-prescriptive way it is looked at by an economist.  If an economist says that they estimate that something will increase GDP, it does not follow that the economist is saying we should do it – they are just reporting a result that this measure of how much is produced will change.  The true question that needs to be ask is whether all the costs and benefits of a policy net out in a way that society would prefer.

A ‘happiness machine’ is closer to the type of thinking that economists come up with – sometimes you may hear economists talk about targeting utility or happiness in this way.  There are inherent trade-offs between different outcomes from a policy choice, and we would like to pick a policy that maximises subjective happiness.  This is a neat way of pointing to, or motivating research into, trade-offs. 

In their deepest dreams and fantasies, economists are utilitarians (maximisers of social welfare or wellbeing) with perfect information about people.  If this was the case, then perhaps we may be able to set up these machines of state – however, in reality there are limits.

The limits of these machines

But this is where economists have become aware of their limitations – the inherent limits of our potential knowledge around matters that are subjective.  We can’t feel what other people feel, we can’t experience things through their eyes, we can’t measure their experience.

In this way prices and markets have a role beyond directly allocating scarce resources and providing incentives to individuals, they also provide information and ensure that people “reveal” their preferences. 

This implies that part of economist’s preference for markets comes through the idea that preferences can be revealed.  To get a wee bit wonkish, in this way we can gain information about the ordinal ranking people place on different outcomes – but even here we don’t have enough information to talk about how people prefer things in a cardinal sense.

Namely, we can often tell that someone may prefer something to something else (ordinal) – but this does not mean we can tell how much more they prefer something (cardinal).  Making things even more difficult, we cannot really objectively compare the value different individuals place on these pains and pleasures!

Markets and the existence of choice allow people to do things that make them happier, but even when they are doing this we don’t have enough information for a happiness index to tell us exactly what to do. 

As a result, a happiness index that is created by an institution or used by a government is merely a representation of what the people involved in making the index see as costs and benefits.  Therefore, it is not clear that an improvement in a happiness index actually represents an increase in wellbeing!

Early economists, especially when the discipline was known as political economy, dreamed of a day when we could measure pains and pleasures objectively.  Sometimes when discussing policy, we can act as if we are able to make up some aggregated indices that are close enough to allow this.  But it is fundamentally not the case, and it ignores the large and persistent gap in our knowledge which should put a constraint on changing the voluntary choices of individuals.

Are you trying to ban government?

This is not to say we shouldn’t do anything – as I noted last week government’s role in helping us co-ordinate is key.  However, the existence of fundamental trade-offs and the extreme limitations of our ability to compare subjective preferences in individuals suggest that we need a different focus when we think about policy.

Instead of targeting an arbitrary set of outputs that treat New Zealand like a machine, policy should be based on the inherent trade-offs that exist for the policy question we are asking.  Focused research on the costs and benefits of educational achievement, health outcomes, benefit policy etc – these are the ways we can incrementally improve policy, and build a better society together.

These outputs may suggest to us there is an issue that deserves investigations– but they should not be seen as an end to themselves.

Policy justified on the basis of the target of an arbitrary GDP or happiness index doesn’t do this, and instead threatens to tie our outcomes closer and closer to someone else’s view of what is right, what is just, what is happiness, and what wellbeing.  Instead the aim of government policy should be to ensure people in society have the ability to reach, and access to, choices that allow them to gain wellbeing.