We could be getting a lot more out of our midwifery services if they were organised differently

Through a rather obscure chain of events, my first job after university involved me in setting the price of cheese across Guyana, a colourful South American country that borders Brazil and Venezuela.

This was before the fall of communism and Guyana was one of the last true believers in a command economy - over four-fifths of its output was state owned and controlled. The government had decided that self-sufficiency was the key to its prosperity, so there were to be no dealings with the external private sector.

If, for example, cheese was to be sold, it would have to be made in new state-run cheese plants. And so at the age of 22 I scaled the commanding heights and was placed in charge of cheese production, alongside my other responsibilities for black-eyed beans, tilapia fish and fuel ethanol.

Of course, before very long it was clear that these attempts at central planning were not only flawed but utterly futile. Whereas state-produced cheese was costly and in short supply, the pavements of Georgetown (the capital city) were laden with black market supplies of cheese and other goods imported illegally from Trinidad and Jamaica. A scene worthy of the Woody Allen film Bananas. Or, one strangely analogous to the weird and dysfunctional world of NHS maternity services planning. Let me explain.

Wind the clock back just three years, when we were told the system of central planning for obstetricians had grossly miscalculated, so that there was about to be mass unemployment of hundreds of obstetric registrars (lo and behold they were then almost instantly absorbed by professionally mandated but efficiency-reducing changes to doctors' working patterns).

Now it's deja vu as we are told there are surplus midwifery trainees who are facing unemployment. And so within the past few weeks we have been informed that in order to meet the desirable new promise of a 'named midwife for every new mother' we allegedly need another 5,000 midwives. Ignoring the fact that the named midwife commitment is reminiscent of John Major's Patients Charter named-nurse pledge, this does all seem a little strange.

As a reminder, last year there were 18,833 full-time equivalent midwives working for the English NHS. And there were 609,300 births. A back-of-the-envelope calculation suggests that this only amounts to 32 deliveries per midwife per year. That's not even one delivery a week for each midwife, even taking account of annual leave and professional training days. And in practice it is even less than that because not all babies are delivered by midwives.

Indeed, perhaps counter-intuitively given the reductions in doctors' working hours, the proportion of babies delivered by midwives has actually fallen from three-quarters in 1990 to under two-thirds today.

Now I know that to even raise these questions is sufficient to get you lynched. So for the avoidance of doubt let me state that I believe a) what midwives do is highly valuable; b) they should have more autonomy; and c) there is more to midwifery than being in attendance at the birth.

Indeed there are the antenatal classes (though many of them seem to be run by the National Childbirth Trust), there are the breastfeeding clinics, the night shifts and out-of-hours care and the postnatal care to boot (even if the division of labour with health visitors is somewhat arbitrary).

But the fact remains that the typical mother doesn't seem to see her midwife for a week or so in total, be it before, during or after childbirth. So it is hard to escape the hunch that we could be getting a lot more out of our midwifery services if they were organised differently.

And now we get to the crux of the matter. Is it another central blueprint-cum-national service framework we are lacking on maternity service reconfiguration or professional demarcations? Or would we do better if midwifery services were in some way self-organising around the needs and choices of mothers, with midwives enjoying high autonomy and strong incentives based on their responsiveness, perhaps similar to the Dutch system?

This is a specific instance of a bigger question that my colleague Professor Julian Le Grand addresses in his new gem of a book: The Other Invisible Hand: delivering public services through choice and competition.

For NHS managers interested in understanding the policy motivations behind recent NHS and education reforms, this is a tightly distilled and closely argued explanation for why these approaches are being deployed.

One of the points that Professor Le Grand demonstrates is that 'voice' mechanisms (such as maternity service liaison committees) will tend to be less effective - and more middle-class dominated - than 'choice' mechanisms (such as giving prospective parents the ability to select where to have their child).

One of my most powerful experiences of this was when my partner and I attended a new parents' evening at our local hospital and came across a large number of Somali couples from east London. They preferred St Thomas' over the Royal London and for once were able to make their preferences stick.

So what we now need is to make sure that these types of maternity choices are extended to include choice of pain relief, delivery method and antenatal care, with money following the patient as against being locked in via block contracts. And then to allow midwives to opt into self-managing co-ops paid on a maternity tariff in accordance with the number of women who choose them.

The Dutch manage a version of this. My feeling is that if we did too, productivity and responsiveness would both improve. Or at least that's a prediction my former colleagues in Guyana would now wholeheartedly endorse.

To see a reader's response to this column, click here