The equity-efficiency trade-off, what’s the point?

Sorry about the constant methodology talk lately, rauparaha and me have just have methodology on the mind 😉

I wish to focus on a subject that is a little different than the rationality definition that my esteemed colleague has been looking at. I plan to look at the equity-efficiency trade-off.

The trade-off between equity and efficiency is one of the primary lessons of first-year public economics. It takes two concepts and illustrates how they behave in the framework of scarcity – which is where the trade-off comes from. As long as we have scarce resources there will be an efficiency-equity trade-off.

However, digging a little further, we might ask why does equity matter? A potential answer is that people value equity, and thereby it is something that needs to be taken into account (something that was described here). But if we care about equity because individuals value it, then why can’t we just introduce equity into the individuals utility function – then the optimal choice will automatically take into account the trade-off between our original idea of efficiency and equity and the new solution will simply be – ‘efficient’.

Is there a reason why we are so keen to divide efficiency and equity – and what does this tell us about economics?

In the efficiency equity trade-off we are discussing two types of efficiency – productive and allocative. Economists assume that the determination of productive efficiency is exogenous from their models, so this leaves us with allocative efficiency to work on. Allocative efficiency is meet when in the market price=marginal cost. There are a number of factors that make us deviate from this – however looking in a broader public economics sense we are only interested in ‘intervening’ when the cost of solving such deviations is less than the benefit.

If we focus on allocative efficiency solely in technical terms, economists feel as though they are providing an ‘objective’ analysis. By looking only at the allocation of resources and not the inherent value that people may place on certain aspects of it economists can define something as efficient.

Equity is a value-laden concept, it is the determination of what is fair. Although we might discover that the allocation of resources is efficient somewhere, it may not be socially preferable as the allocation is unfair. In some sense this is related to the idea of distributive justice – and can be proxied by way of an externality.

So the idea of equity is normative – ergo why economists ignore it and focus only on the aspects of ‘efficiency’.

However, over time this trade-off has become confused. The idea of externalities and social values has allowed economists to also use the term efficiency when describing the socially optimal solution. This is a problem.

If economists believe that the ‘efficient’ solution is the optimal solution, it can give rise to people ignoring the effects of equity – thereby making an implicit value judgments that the equity impacts are equal to zero. This is of course false.

The economists of yesteryear had a good point when they split up objective measures of efficiency from subjective measures of equity, and it is a point we economist can forget only at our peril.

However, should we continue using our method to only derive efficiency? Can we not provide a framework that looks at both efficiency and equity, but leaves the derivation of parameters (the value judgments) to policy analysts? Sure we only be able to quantify efficiency, but our framework is a lot more powerful than that – I would like to hear your thoughts.

9 replies
  1. CPW
    CPW says:

    “This is of course false”, sounds like someone got up on the normative side of the bed…

    Actually, if we expand our equity concerns to future generations, and throw in a low discount rate, I’m certain you could get situations where the efficient solution is optimal, even when equity does enter into the welfare function. Just sayin’, that’s all…

  2. Matt Nolan
    Matt Nolan says:

    “if we expand our equity concerns to future generations, and throw in a low discount rate, I’m certain you could get situations where the efficient solution is optimal”

    The discount rate is subjective 😉

    In that line “This is of course false” I’m not saying that we can’t do a lot with our tools, I’m just saying (again) that optimal policy comes from value judgments, and we have to realise that our analysis only covers one part of that.

    My main question has to do with how we set up our models. Should we include equity considerations in our framework but then not define the parameters, or should we focus on only looking at efficiency and completely ignoring equity in our models. Or can we do both – as long as we are clear about the difference. I’d go for the last option.

  3. CPW
    CPW says:

    Some of the happiness research also suggests that country level GDP correlates with happiness, but inequality doesn’t, which would again suggest that a plausible parameterization (on that metric at least) would result in efficient=optimal.

    Policy decisions include equity decisions, does explicitly modeling them within the model without endorsing any particular equity function add anything? I can’t think of how there is any practical difference here. As i said the the other day, on a lot of policy issues the equity issues are external to the analysis, depending on what the opportunity cost is.

  4. Matt Nolan
    Matt Nolan says:

    “Some of the happiness research also suggests that country level GDP correlates with happiness, but inequality doesn’t, which would again suggest that a plausible parameterization (on that metric at least) would result in efficient=optimal.”

    I think we might be talking cross-purposes here. I agree that efficient can equal optimal. However, the efficiency-equity trade-off that first years learn involves separating the objective measures of efficiency from the bits that require value judgments – which we then term equity.

    In reality I guess I have a semantic problem with efficiency being used for two different things.

    “Policy decisions include equity decisions, does explicitly modeling them within the model without endorsing any particular equity function add anything?”

    Good question. I think it only adds something insofar as the economic method might be a better way of framing those issues than other methods. If an economist, with all there experience making models, can provide a general model, which policy analysts can then set values into I think we could be in a good place.

    Also note that I separate the role of policy analysts and economist, however these don’t have to be separate people. The important thing is that the two steps are recognised as separate.

  5. CPW
    CPW says:

    Sorry, I was just pedantically observing that the idea of an equity-efficiency trade-off is itself subjective.

  6. Matt Nolan
    Matt Nolan says:

    Don’t say sorry for stuff CPW, you haven’t done anything wrong 😛

    “the idea of an equity-efficiency trade-off is itself subjective”

    Statistics on ‘efficiency’ in technical terms are objective, while equity is subjective.

    Anything that defines a trade-off between a objective and subjective element must be subjective.

    So are you saying that the method by which we introduce equity into our models is subjective (as we have to define where to put it in the individuals value function). I would agree with that, I think that is a very appropriate point.

  7. Oliver Woods
    Oliver Woods says:

    Hi guys,

    I’ve made a post over at my blog that uses part of this post in the discussion. You might be interested to check it out! 🙂

    Cheers,
    Oliver

Trackbacks & Pingbacks

  1. […] economists are always right… or wrong 5 03 2008 Oliver Woods has used Matt’s post on the trade off between equity and efficiency to launch an attack on the separation between […]

  2. […] tales of taxing talent 2 03 2008 Matt’s post on equity and efficiency reminds me of a paper by Greg Mankiw and Matthew Weinzierl on optimal […]

Comments are closed.