Tuesday, December 16, 2008

Limitations of Economics

In a previous post - The difference between economists and non economists, I gave the impression economists are a paragon of virtue, rationality and common sense amidst a sea of ignorance, superstition and irrationality. Some have suggested the post is condescending and patronising. They are probably right, but, sometimes it is good to state a case in strong terms, to make people think. However, I feel there is a need to redress the balance and point out the many mistakes / limitations of economists.

Economics is difficult

I think it was Keynes who said economics is very difficult and many people underestimate how difficult it is. In maths 2+2 always =4. But, in economics it usually depends on countless variables almost too difficult to take into account. To give one example, the link between the Money supply and inflation. The quantity theory of money MV=PY suggests there is a correlation between the money supply and inflation. (as most non economists would tell you - if you print money you will cause inflation) But, in practise the growth of the money supply is influenced by so many variables such as technological change, velocity of circulation and consumer behaviour that M3 growth statistics became almost meaningless. - MV=PY is great in theory but in practise it is difficult to make anything out of it. (money supply and inflation)

Forecasting the Future

It is difficult to forecast the future; yet in economic policy making, it becomes important . Go back to May 2007 and how many economists were predicting a fall in UK house prices of 25% and the deepest recession since the war? I wasn't and certainly not the treasury economists, who were predicting stable growth of 2% and a reduction in the government's borrowing. Of course, there were people predicting a house price collapse and they have been proved right. (Though some of them started predicting a house price collapse back in 2000.)

Difficult in Knowing where you are.

One of the great challenges is knowing the current state of the economy. For example, China's growth and unemployment figures are always viewed with suspicion. There is great debate about what the US inflation rate is - it depends which model you use. Recently, the US GDP statistics were revised meaning that the economy was in recession much earlier than previously thought. How can you make good policy when you don't even know what happened in the past? - let alone predict the future.

Using Old Models

In a way, this recession is unusual in that it wasn't preceded by an inflationary boom. The government felt that as long as inflation is under control, the economy must be sustainable. However, the mistake was to ignore an asset and lending bubble. The problem is that it is not sufficient to rely on previous experiences. As the economy develops old models become less relevant.


A good economist would be free from ideology and have a willingness to revise theory in light of empirical evidence that doesn't match upto their beliefs or expectations. However, in practise many dislike evidence which doesn't agree with their point of view. For example, some economists place great faith in the virtues of the free market and therefore take a lazy attitude in assuming free markets will always lead to increase economic welfare. The problem is that free markets can often be beneficial. But, at the same time, there will always be exceptions; you can't make generalisations that free markets are always best nor can you make generalisations that free markets are always wrong. It is necessary to evaluate each on a case by case basis.

Another example of ideological economics could be the assumption that if privatisation works in one country it must be good for other countries too. Free trade is another example. Most economists will tell you free trade is beneficial. But, this doesn't necessarily mean developing countries should always stick to the free trade mantra. - There can be exceptions to every rules, for example the infant industry argument.

No comments: