All entries for Monday 23 February 2009

New Romantics

Writing about web page http://www.guardian.co.uk/commentisfree/2009/feb/20/economics-emotions-human-values

There is a strong case for thinking about how emotion and mood affect economic decisions.

Does my own mood affect my decisions? I discussed this with my wife, and she agreed I'm a model of level headed rationality. But she knows lots of people for whom that wouldn't be true. She's not saying who (but I bet they're not economists).

If mood can affect decisions, does it affect the decisions that really matter? Not all decisions are important. But isn't it at least possible that powerful emotions like joy, fear, sadness, or enthusiasm interfere with our ability to calculate an optimum? Suppose that emotions frame our vision of the future; suppose they are capable of boosting our willingness to provide for the future or winding it down; suppose they make us more or less willing to shoulder risks. In that case, important economic decisions will indeed swing with our moods.

Again the net effect, averaged across millions people, might not add up to much. If one person's mood cancels out another's, the total effect should be zero. But if the moods of millions swing together, in a concerted way, billiions of pounds could swing with them into -- or out of --  particular markets.

And it is intuitively plausible, to say the least, that the mood of millions of people is taking a hit at the moment, as homes and jobs are lost and the fear of loss infects our nation.

As Jon Elster pointed out some years ago in The Journal of Economic Literature (1998), economists have given much closer attention to cognitive limits on rationality than to emotional limits.

Which brings us to the critics of orthodox economics. There are so many at the moment ... it seems invidious to pick and choose. But choose we must, so we'll pick from my daily newspaper which, despite the fact that it is written mainly by and for lunatics, remains The Guardian. On February 16, 2009, Larry Elliott wrote:

There have been many economists down the years who have expressed scepticism about reducing their discipline to a mechanistic subject. Malthus told Ricardo to be wary of becoming too attached to abstract hypotheses; Schumpeter talked of creative destruction; Hayek saw the market as a voyage of discovery; Keynes stressed the importance of "animal spirits."

And:

Mervyn King says Britain is in a deep recession ... Interestingly, the governor cited Keynes at the Bank's inflation report press conference, noting that animal spirits were currently depressed. With confidence so weak, it is hard to envisage an early or a robust recovery.

Two days later, Sam Whimster (Professor of Sociology at London Metropolitan), commented: 

We should also consider the place of emotions in economic life. The share price of UK banks fluctuates wildly as traders attempt to calculate their capital value from future estimated losses and profits. Keynes, in 1933 in his lectures on his General Theory, said that current yields of firms exercise an "irrational" influence on estimating future worth.

In the boom years that have ended so abruptly, Whimster continues:

Infectious greed and optimism was the mindset of economists, bankers, politicians and regulators - leading to behaviour that no regulatory mechanism could have controlled. But the extent of the greed and adventurism, and the flouting of standard banking precautions which had been stress-tested by decades of history, raises the question of what determines which emotions come to the fore.

The question of whether economists should take emotions into account is a good one. But where are the answers? Not as easily to hand as one might hope. Let me mention some issues that everyone should think about at this point.

To begin with, emotions are just one more variable. There is a lot going on in the world economy that we understand all too imperfectly. But it doesn't help our understanding to say, after the event: Oh -- people have been behaving irrationally, it must be because of their emotions. Change the context a little and you'll see how fainthearted and pathetic this is. Imagine me telling my wife: You've been behaving irrationally, it must be because of your emotions. (I can't imagine saying it, but you can try.) She'd kill me, and rightly so. The reason is not just that it's insufferably patronising, but that it also devalues emotions into something irrational, flighty, whimsical, and beyond understanding.

In different words, adding another variable, the variable of emotions, to a model does not add to its predictive power unless the added variable is itself predictable. Are emotions predictably variable?  It is entirely possible that aggregate moods are predictable -- in fact, I made a prediction above when I suggested that "that the mood of millions is taking a hit at the moment, as homes and jobs are lost and the fear of loss infects our nation." If so, perhaps we should be working towards a model of emotions. But be aware that it may not be worthwhile. If the aggregate mood just goes up and down with the rate of unemployment, for example, then modelling emotions may add complexity without increasing the predictive power that is already taken there in conventional models.

A deeper question is whether emotions truly vary in ways that are beyond our individual control. Consider hatred. Two sorts of people can kill without hating. Some are trained and paid to do so to defend us against our enemies, and these are our soldiers. The rest are few in number and we call them psychopaths. The majority of people kill only those whom they hate. So, hate first, then kill. But here there is a problem: the idea that killing, or any kind of premeditated violence, is the deterministic result of uncontrollable feelings, flies in the face of our traditions of personal moral and legal responsiblity. In fact, it is possible that we must first allow ourselves to hate those whom we have a prior intention to kill. Decide to kill, then hate, then kill. If so, emotions as such do not decide what we do; we decide what to do, and then select the emotions that validate our decisions. This is a difficulty that must be resolved before we can understand whether emotions add anything to economics.

Clearly, we must learn more about the psychological laws underlying emotions. I am open minded as to whether we will ultimately need more complex economic models. Whatever we decide, I am in favour of economic models -- maybe not more complex ones, but just better ones.

If so, I am going to disappoint Elliott, in whose view:

The mechanistic approach to economics has failed.

"Mechanistic" is a prejudicial term in this context, however. Economic models are just thought experiments. They are mechanical in the sense that they are formal (as opposed to purely literary and intuitive) and logical (as opposed to internally inconsistent). They still seem like useful things to me. For argument, I will quote briefly from the best essay written on this subject in the last 30 years: Paul Krugman's "Two Cheers for Formalism," published in The Economic Journal (1998); I also strongly urge you to read the original in full:

First, much of the criticism of formalism in economics is an attack on a straw man: the reality of what good economists do is a lot less formalistic than the popular image. Bad economists, of course, do bad economics; but one should not confuse a complaint about quality with a complaint about methodology. Second, when outsiders criticize formalism in economics, their real complaint is often not about method but about content - in particular, they dislike "formalistic" arguments not because they are formalistic, but because they refute their pet doctrines. Finally, as a practical matter formalism is crucial to progress in economic thought - even when it turns out that the ideas initially developed with the help of formal analysis can in the end, with some work, be expressed in plain English. Moreover, this is especially true precisely in the sorts of areas that economists are often accused of neglecting, such as those that involve imperfect competition, incomplete rationality, and so on.

I'm not sure where I should leave Whimster, who starts his comment by asking to bring in emotions and ends up bringing in values; the problem, he concludes, has been that:

Anglo-Saxon attitudes have been dominated by what Weber would have called the values of adventurer capitalism, and the economist and sociologist Werner Sombart would have called the lust for wealth ... Aesthetics, harmony with nature, the ethics and politics of community - these need to be reasserted as values independent of and superior to market values, which as the Romantics pointed out should be merely means to other ends. It is time for what Nietzsche termed the revaluation of values.

Don't you love the wedge that's driven between "market values" and higher things? We're losing a hundred thousand jobs a month, but it's okay because we should refocus on "aesthetics, harmony with nature, the ethics and politics of community."

Markets that work well have allowed millions of people to live and work as they choose, enjoy the freedoms of the press and media, travel the world, and address the world from their own homes. Markets that work well have another virtue, as Hayek understood so clearly: they allow one person's end to be another person's means, without anyone having to rule on or even consider the difference. Millions of people aspire to what well-functioning markets can offer: opportunity, mobility, prosperity, stability, insurance, the backdrop to the everyday joys of friendship, love, sex, children, and families -- or not, if you don't want them. I don't know what's higher than that.

That's why, when markets don't work well, it's so important to fix them.

Mark Harrison : 23 Feb 2009 22:56 |  Tags: Keynes Politics Recession |  Comments (0) |  Close comments |  Report a problem