AFTER CHRISTMAS, the hangover. Good cheer will be in short supply in the new year for the BBC, the NHS and many other organisations in both public and private sectors that face debilitating rounds of cost-cutting.

This column has a better suggestion. Instead of saving pennies by shaving routine expenditure and delaying payments to suppliers, why not resolve to do things better – not just a bit better, but orders of magnitude better so much better, in fact, that they become cheaper?

No, I haven’t been at the port again. Economists will tell you that cost-cutting becomes progressively more difficult and expensive for every extra unit of resource saved, and they are right – if every increment is saved the same way as before. However, large savings in resources can sometimes be cheaper to make than small ones. This has been labelled ‘tunnelling through the cost barrier’ by Amory Lovins of the Rocky Mountain Institute (RMI), and it works like this.

If you build a house, the conventional idea is that the more energy-efficient it is, the more it will cost (double-glazing, thicker insulation). So a moderately more efficient house will indeed come out dearer, and diminishing returns mean that after a certain point any additional gains will be outweighed by the increasing cost of the improvements.

But this isn’t the end of the story. Suppose you install superwindows, superinsulation and superefficient appliances. Each one in itself is un-cost-effective. But if together they eliminate the need for central heating and air conditioning , they may actually reduce the total upfront investment. So the more ambitious house is cheaper to build, as well as saving much more in running costs.

Why is the idea that better means more expensive so deeply ingrained? Because people almost always look at design elements in isolation. As Lovins notes: ‘You can actually make a system less efficient while making each of its parts more efficient, simply by not properly linking up those components… Optimising components in isolation tends to pessimise the whole system – and hence the bottom line.’

What applies to engineering applies to business as well. The key is designing the process as a whole system. In manufacturing, long before the advent of ‘lean’, Richard Schonberger christened as ‘frugal manufacturing’ the dynamic by which, if they are systematised, sav ings multiply each other (the real definition of synergy).

He described it as the ‘pursuit of the zeros’: zero waste time, effort, movement between processing stages, work in progress and, above all, product no one wants. The shorter the time between order and delivery, and the more closely coupled the processes, the more resource-efficient the system – and, just as the well-designed house needs less heating and cooling, the well-designed factory needs less computer power for scheduling and planning, and less management to run it. Better is cheaper.

All this applies to services, too. Cost-cutting by decimating headcount at contact centres or back offices is laborious, costly, and on its own does nothing to improve service. Instead, the aim should be to redesign the process, as a whole system for the best outcome and then decide on the scale of inputs needed.

Yet the reverse is the general rule. For example, the government wants local authorities to use specialised document-processing IT to speed up the payment of benefits. But careful analysis shows that the principal cause of delay in benefits processing is not the back office but the front – getting ‘clean’ information from claimants that agents can act on. Councils that concentrate on the front end consistently clear claims in a week, instead of the 50-60 days that are standard, and barely need computers at all.

Or consider banks’ contact centres. These were set up with the idea that mass-production telephone factories would cut costs by processing calls faster than branches. But by distancing customers from branches, contact centres multiply the number of queries, creating the need for yet more contact centres, all answering questions that shouldn’t need asking in the first place.

Now the government is making the same mistake on an even grander scale by mandating shared-service centres for groups of local councils or departments. But economies of scale are irrelevant at component level. This was discovered the hard way by a US aero-engine manufacturer that reorganised production around a bank of state-of-the-art automated grinding machines for turbine blades. The machines were so fast that they slowed the system as a whole to a crawl, clogging it with huge batches and requiring expensive technicians to programme them. Replacing them with slower, simpler machines meant that grinding took longer, but the entire production process worked many times faster – and the cost was halved.

So cheer up. The hardest part is getting the brain around the idea that, for once, there is a free lunch.

The Observer, 31 December 2006