IN the era of quill pens, paper, advice notes, invoices and ledgers, double entry book keeping evolved to provide a combination of checks and balances to minimise human error and fraud. A well-kept ledger was a picture of financial clarity for a world meandering along at a snail's pace. Today communication networks link cash tills, points of sale and transaction, production lines, logistic chains, banks and people. Everything from raw material procurement, to inventory control and customer support is now online. The ledger is long gone and computer spreadsheets and graphical presentations have taken over - and are capable of much more.

The concept of the financial year and exactly how it originated seems to be lost in history, but chances are it is related to the gathering of taxes. Certainly, the arrival of the ledger and double entry book keeping are part of the same mechanistic need for people to see how much and where the money is when. But it is as if we have not moved on. In every country we find companies, banks, accountants, legal institutions, markets and governments still abiding by the financial year.

In most companies and institutions the financial year promotes many strange and unnatural acts. Money has to be spent by some magic date, things have to be shipped and invoiced, a line has to be drawn, accounts have to be in. Why?

It makes no sense. Surely we can just grab the numbers as they flash past on April 5 each year. Why do we continue to switch entire financial systems on and off abruptly? In a world that is speeding up and moving ever faster, the result of such action can only be to introduce transients into the management systems of the organisation. In a world now dominated by bits and not atoms, such self-induced transients can take weeks to settle down, and the resulting wounds can be far more damaging than those inflicted by a fiercely competitive market.

If you had a child on a swing eating sweets, you wouldn't stop the swing abruptly to check on the consumption in some arbitrary period. Think of the risk involved: the child could fall to the ground, be hurt and scatter the sweets in the grass, and then be unable to recover them all. Even worse, a sweet might get trapped in the throat and the child could choke. Much better to stand on the sidelines and count the sweets as they pass the lips. We may make a minor error or two in the counting process, but it would be far safer, and sufficiently accurate.

Most dynamic systems are like the child on a swing; they just don't like abrupt starts and stops. We would never consider stopping a steel works, oil refinery, power station, or manufacturing production line at the end of the financial year, only to start them up again the next morning. Economically and operationally it would be a disaster. But people actually do this in large and small companies. The entire operation is disrupted by a sudden brick wall stop, followed by an attempt at an instantaneous restart. It is about as sensible as closing the store for stocktaking and losing a full day's trading as a result.

Peter Cochrane holds the Collier Chair for the Public Understanding of Science & Technology at the University of Bristol. His home page is:
http://cochrane.org.uk