Over the next several months, therefore, this column will argue that economics has comprehensively lost its way. It will take the help of the latest economics research in order to do so.
It has become commonplace to say that economists never agree on anything. But even if the number of disagreements is directly proportional to the number of economists, surely there must be some common ground?
But, like the Brahmins of the Gupta period — one of whom discovered zero, whence the name of this column — economists are today obsessed with empirical method, rather than substance. This may not have mattered much as long as those who do bother with substance paid some attention to the method of logic. But, often, that too seems not to be the case. The result is a discipline in which the only certainties are the ones in microeconomics postulated by economists in the 19th and up to the mid-20th century.
Why empty sum?
I have deliberately chosen ‘Empty Sum’ as the name of this column. It sounds very similar to the better-known concept of ‘zero-sum’ from Game Theory. But the latter has a very specific meaning and signifies the trade-off where neither party to the trade either gains anything or loses anything. Empty sum, in contrast, means that zero multiplied by zero will give you zero.
You only need to look at Bramhagupta’s rules about zero to see how closely they fit economics. And, today, as with zero, the problem with economics remains what the ancient Greeks grappled with: What do we do with it? Can something be a number and not a number at the same time?
Economists will protest, more than somewhat shrilly, that I am making far too many sweeping statements. That is to be expected. But, after spending a lifetime with them, I know that in their heart of hearts they agree that this is exactly true of most of post-war economics.
Techniques in econometrics have, of course, improved. But useful economic theory has made virtually no progress. Much of microeconomics stays where it was in, say, 1945, except in the matter of its dress. And as we are seeing now, macroeconomics hasn’t gone beyond Keynes, who published his seminal work in 1936.
My view is that the post-war period can be divided into two clear segments. The first comes roughly up to the mid-1970s when economic theory grew increasingly distant from real life and more-or-less merged into advanced mathematics. The second phase was a sort of reaction, and comprises the empiricist phase where, gradually, the mere presence of a data set has meant that anyone who knows a little regression can become an economist. Thought and thinking have both exited.
The trouble with empiricists, as opposed to empiricism whose importance can’t be denied, is that they can’t tell the difference between correlation and causality. I once wrote an article showing how every change of party of government in the US after 1945 has been accompanied by a recession.
The correlation was easy to establish but causation, in either direction, is another kettle of fish altogether. Even after having been advised early not to fall prey to it, they make the classic post hoc, ergo propter hoc (after this, therefore because of this) mistake.
Swings between extremes
The triumph of mindless empiricism has seen the advent of a menace called the policy economist. This tribe is to economics what ‘non-state actors’ are to security: guided missiles but with commercial intent. These economists flourish in think tanks which, if you ask me, are like those camps in Kandahar, Muridke, PoK, etc. But more about them in the months to come.
Overall, the swings between the two extremes — of economics posturing as physics so that it can use advanced mathematics, and data-based vacuity which has no roots in politics, custom, institutions, law, etc — have led the discipline into a state of meaningless tarkam and vitarkam where scoring petty points is mistaken for scholarship and where debate becomes an end in itself.
Let me conclude by saying that Oscar Wilde, who said that an economist is a person who knows the price of everything but the value of none, was wrong. Today, it is the other way round.
6 months ago