Home > Politics > Positive and Normative Economics

Positive and Normative Economics

Hi all,

I have not written here for a long time. Long story short, I have left London to return to Australia, and have started a university degree in politics. This piece is addresses to the students in the introductory microeconomics class that I am taking. But it may be of interest to anyone who is concerned with economics or is affected by the way that economics is taught (that would be all of you)…

I would like to say a few words about something that has been bothering me through these early weeks of the Microeconomics 1 course. This is the distinction between positive and normative economics. This was touched on briefly in the first week and then swiftly left behind which I think is unfortunate as many of the claims in the lectures and the textbooks that appear on the surface to be positive would also appear to have rich normative content. It may also be the case, as I hope to show, that some claims that are labelled “normative” can be subject to positive methods. The ability to distinguish between positive and normative claims, and detect when one is submerged in, or disguised as, the other is one of the most important critical tools in we have in the social sciences. I am not here arguing that any given norm is right or wrong. Indeed, we cannot even begin to approach such questions until the norms are explicit. While I understand that this is an introductory course, the ideas that we are taught at this stage are internalised and can colour the way we see things going forward. If we do not develop critical skills from the first day we may never do so. We may get good grades by memorising the claims, but we risk becoming bad economists. And I fear that the world has enough of these already. I personally don’t want to be an economist but I do want to live in societies where I can trust those that are to do the job. One of the many critiques to emerge after 2008 is the observation that economics as a profession has been quite poor at making this positive / normative distinction. (Another is that economics had become obsessed with its models to the detriment of dealing with the world as it is, further undermining positive claims). These criticisms have come from notable economists as well as many standing on the outside looking in.

The given distinction between positive and normative economics is roughly that between descriptive “is” claims (empirical claims that can, in principle, be tested against the real world) and prescriptive “should” claims (value based and inherently untestable). Of course, making claims with the word “is” is an insufficient condition for positivity (and this is at the heart of my problem with what we are taught). I could say that “economic claims are 90% normative and 10% positive” – this is probably a normative claim rather than the positive claim that is appears to be (I can think of no way to test it, even if I might feel it to be true). We need to develop a richer feeling for this distinction than a simplistic is/ought dichotomy provides. It helps to have a few (non-exhaustive) examples of similar claims to highlight the difficulties.

In one of the textbooks there is a statement about taxation “distorting” incentives. This appears as a positive claim. This really stood out for me, coming so soon after the brief coverage in the text of positive and normative claims. It seems to me that there are many normative implications hidden in this: what sorts of incentives are acceptable? Why is it a “distortion”? This has a value-loaded tone, and has a well-defined meaning implying a negative change away from a given norm (that word again!). (I would consider it a more positive claim if it stated that taxation alters or modifies or affects incentives, which is uncontroversial but inconclusive as taxation can be used as a positive incentive against certain undesirable behaviours). But more than that, this is quite possibly empirically false. Plausible arguments exist that suggest that, at least at higher income levels where material well-being is increasingly irrelevant, incomes are viewed comparatively against a person’s peers as a kind of score-card and that higher taxation rates do not effect incentives to any great degree. Where this becomes problematic is when different tax regimes compete for such people. But this is not a problem of taxation per se. This is to say nothing of zero-sum positional goods and the like. Another argument is that higher incomes are actually a disincentive (I’m thinking here of recent work by Daniel Pink and others) and that there is little correlation between pay and performance at very high levels. All of this is conceivably positively testable once we bring out the hidden normative content.

Another claim that has been made in both lectures and the texts is that higher productivity equates to higher social welfare. A bigger pie is a better pie. Intuitively we can see some immediate problems here. Part of the difficulty is, of course, that the pie is conventionally measured in money and money is not well-being (nor can it buy you love). I, being exceedingly rich compared to you, can bid you out of a market for something that you need and I merely fancy. This is economically efficient but is it really efficient, more broadly conceived? No amount of relativistic scepticism can remove this question. We might argue that “revealed preferences” are not interpersonally comparable, but using money as a proxy does nothing to alleviate this – it merely defers the problem. Common sense and appeals to norms may be useful in such cases. Measured in money, more bread/medical care/housing/etc provided to people can look less efficient than less of the same goods provided at much higher prices. Ability to buy may be as, or more, important than willingness for many goods. (Imagine the choice in a society of 100 people for the body of bakers – do they produce loaves for 100 able buyers at $1 per loaf, or for 50 willing buyers at $2.10 a loaf – a contrived example, to be sure, but one that has many real-world applications). Ignoring distribution is a common and well-documented problem with theories that are rooted in utilitarian and consequentialist approaches. As productivity continues to exceed need distribution becomes an increasingly important factor. The model of supply and demand profoundly shapes and is shaped by distribution. Beneath this ostensibly positive model lie assumptions and judgements about an underlying property model which is both deeply normative and contentious. Beyond this, there are plausible arguments being made that gains to general well-being diminish sharply with economic growth. This has led several prominent economists and social theorists to start asking what we should be measuring instead of the traditional national accounts. Among other accounts, the distribution of wealth is currently being cited as a possibly important factor. One commonly heard argument is that there is a trade-off between “equity” and “efficiency” where it is strongly implied that equity is a normative notion and efficiency is positive. (Closely related to this is the argument that there is a trade-off between liberty and equality, an argument that rests on strongly normative ideas about liberty). But this supposed trade-off has recently come into strong doubt. Economists such as Joseph Stiglitz, and more recently IMF economists, have suggested strong links between market efficiency and equality or equity. What was presented as a normative versus positive dichotomy appears to have a strong positive component, something that is missed when it is assumed that equity resides solely in the normative zone.

Yet another claim that we have heard, and my final example, is that trade is required for specialisation to work its magic. This is not the case if we are to use precise language. There are two distinctions here that are important. The first is between production and distribution, and the second is between free production and non-free production.

We have heard the examples of Robinson Crusoe and Man Friday on their island. They split the work by specialising according to comparative advantage and achieve significantly higher productivity, and hence welfare gains. Imagine instead that Crusoe arrives at the island with a pistol. He quickly figures out the comparative advantages of himself and Friday. He then commands production according to these comparative advantages. Friday must collect coconuts while Crusoe himself gathers fish (or vice versa). This would result in the same economic outcomes as would be the case had they agreed to produce according to comparative advantage. But this is a very different situation: it is now a command economy. Goods are not “traded” in the way that we conventionally understand the term – they are allocated through diktat. Production is the same, distribution differs. The social outcome is different in certain qualities even if not in quantities. It might (tentatively and with many qualifications that delve deep into norms) be the case that an agreement to divide the work up exhibits what I have called “free-production”. It is certainly the case that the command version of the story represents “unfree production”, at least for Friday. A utopian “Marxist” allocative process might involve people producing according to comparative advantage (from each according to their ability), pooling the output and taking what they need (to each according to their need). Or the pool might be divided, by common agreement, in equal portions. I’m not arguing that this would work well – I am simply trying to draw down the important distinction between production (where specialisation occurs) and distribution. This, again, differs from trade as conventionally understood. Stretching the term “trade” to include these scenarios (e.g. by describing these as “trade agreements”) would require us to describe the former Soviet Union as a predominantly market society. Aside from the ideological loss, we lose descriptive power and, with it, an understanding of what we are describing.

Command economies rely on specialisation at least as much as market economies. It might even be argued that they could, in principle, do so more effectively: imagine that a planning bureau (or whatever) had information about comparative advantage it could guarantee that people produced according to their comparative advantage (there is nothing in command economies that says they could not apply comparative advantage as a measure). Granted, missing the pricing mechanism, people may produce more things, but less of what is needed. Conversely, in a “free-market” people may choose not to engage in the activity that is their comparative advantage. They may prefer to pursue something they enjoy but are less comparatively good at, for example. Truly free choice can mean doing things that are not economically optimal. This may well produce sub-optimal (inefficient) market outcomes but may well produce more efficient social outcomes. In one of the textbooks is an example of a household of two people who divide the housework. The exercise is, of course, designed to demonstrate that they can do more by specialising which is unarguably true. But anyone who has lived with other people will know that this may produce a socially inefficient outcome and that letting people do at least something of which they like doing the most may lead to a happier household even if it the household is a bit messier. This can hold true at the societal level as much as that of the household. Perhaps this distinction is what is wrong in a command economy (and in a market economy where some people have few and unenviable choices): again, true choice can be sub-optimal. These are, to an extent, positive claims inasmuch as social welfare can be measured (and we generally know whether we are living in a happy household or not, even if this is trickier at a macro level). Interestingly, development economists sometimes describe comparative advantage as a trap: doing what I am good at now can preclude becoming better at things that will bring long-term advantages. We get better by doing, and we can only do so by at first doing it badly. Many countries have developed by deliberately choosing things that they were not comparatively good at. South Korea is a good example. Without their rejection of the logic of comparative advantage we would not have Samsung, LG, or Hyundai.

There are more examples that we have encountered, and undoubtedly many more to come. But without being alert to the hidden normative content of claims we are less able to draw out these sorts of implications. If we are aware that apparently positive claims can have some or much normative content and that apparently normative claims can be subject to positive methods then we can all be better economists, or at least better citizens.


Categories: Politics
  1. No comments yet.
  1. No trackbacks yet.

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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: