Too many managers

Before the 2010 election I heard the phrase "too many managers in the NHS" many times. I also heard the "statistic" that in 2009 "the number of managers increased ten times more than the increase in nurses" mentioned often. The right whinge commentators at Conservative Home said that this was because Labour supporters were taking high paid jobs while they still could. Nonsense, of course, but it is a meme that sits in the mind of many Tory supporters. Of course the Tory commentators did not also say that the number of managers went down by -7% in 2006. Nor did they say that the amount of work carried out by the NHS went up by 65% and the total number of employees went up by 35% between 1997 and 2009. (More work, more employees, means more managers are needed.)

Totally out of the blue, the Draft Manifesto of the Conservative manifesto said that administration costs in the NHS had to be cut by 30%. It gave very little idea how this could be done other than waffling about an "information revolution". As a software developer (and trainer, and writer) I am all for yet another "information revolution", but I cannot see how it will cut administration by 30%. The majority of NHS administration is in stuff that cannot be cut by such a large amount: appointments, scheduling diagnostic tests, use of facilities, and supporting the "internal market" where every procedure has to be billed to someone. I am sure that savings can be made, but 30% is just castles in the sky.

Then after the election Lansley had a brainwave. SHAs and PCTs cost £1.85bn a year, if their work is transferred to the private sector at a cut price rate, then the election pledge can be kept. In the NHS white paper Lansley said that there would be a 45% cut in management through the decision to abolish SHAs and PCTs, so that commissioning would cost £1bn. I'll come back to some of the actual costs of this decision in a later blog post.

Recently the phrase "the number of managers in the NHS doubled under Labour" came to my attention, so I thought I would check it out. The first place to look is the NHS Information Centre. However, these figures only go back to 1995 and I wanted to compare the Labour terms with the preceding Tory terms. Another source of information is Hansard. MPs have a habit of asking questions about NHS statistics, sometimes asking the bizarrest format. Anyway I was able to find one question in 1993 asking for the numbers of administrators and managers since 1975, and this gave figures up to 1991. I still had three years to fill in and I was able to get 1992 and 1993 from another Hansard question and the remaining figure (1994) came from a graph in a report by the Kings Fund (Figure 36).

Putting all of this information together I was able to plot a graph of the number of NHS managers since 1985.

What is immediately apparent from this is that the rate of increase in the number of managers actually slowed under Labour. From 1988 to 1994 there was a sharp increase, year-on-year, in the number of managers. There was a drop in 1995 and then a slower increase until 2002 when the rate increased again. The rate of increase between 2007 and 2009 was comparable with the Tory rate between 1988 and 1994.

This graph shows that increasing managers is a Tory phenomenon as much as it is a New Labour one. So the sentiment that "the number of managers in the NHS doubled under Labour" and the implication that a Tory government is needed to get this under control is hypocritical at best.

Lansley says that he wants to cut management by 45% (by about 20, 000). This will take us to the 2000 figure at around 25,000. Is this possible? Few people think that it is. However, I think that it is perfectly possible for one important reason: the graph shows how many NHS managers there will be. Lansley wants to privatise commissioning and this means that manager NHS commissioners (who are classed as "managers") will lose their jobs. The work will move to the private sector where there will be no public figures of how many managers/commissioners will be used. Lansley will get the drop in the graph he wants, but the graph from that point onwards will show a different measurement entirely.