³ÉÈË¿ìÊÖ

Wednesday's GDP figure for the last three months of 2011 will be the most eagerly-awaited economic data of the year so far.

In recent quarters we have been surprised by a few GDP figures, but should we have been?

We know what goes into the economic growth figures, and we already have some of them for the first two months of the quarter.

The first estimate of GDP is based on the production approach, which involves the Office for National Statistics (ONS) adding up all the value created by the production of goods and services in the country.

The ONS has measures of the amount of activity in the manufacturing sector, the service sector and the construction sector, which it uses to tell us whether the economy has grown or contracted between quarters.

The problem is that for the preliminary estimate of GDP, the one that is coming out on Wednesday, the ONS is taking an educated guess at a figure, based on knowing about approximately 40% of the economic activity that will be included in revised estimates made over the next year.

There is a trade-off between getting a preliminary figure for GDP out as quickly as possible and including as much actual data as possible.

Bearing that in mind, it is perhaps surprising that there are not greater revisions to the GDP figures.

So, let's take a look at the information we have so far to take an educated guess.

We know what sort of figures the ONS has for October, because they have already been published.

In October, the declined 0.7% compared with September, due to contractions in the transport, storage and communication category and also in business services and finance.

That's a bad start, because the service sector accounts for more than 70% of the UK economy, so if the service sector is contracting significantly it is very hard for the GDP figure as a whole to be positive.

The Index of Production fell 1% in October, compared with September and then another 0.6% in November compared with October, so manufacturing was looking pretty bad.

The third factor is the construction industry, which the ONS said declined 2.5% in October compared with September and then rose slightly, by 0.2% from October to November.

So the key figure that is missing is the index of services for November, which will be released along with the GDP figures on Wednesday.

We also do not know what happened to any of the three indicators in December, so we have to look at other indicators.

At the end of November, : "We expect GDP to fall slightly in the final quarter of 2011."

That goes along with an average forecast of -0.1% from analysts polled by Reuters this week.

There has also been much discussion in the media about whether it makes much difference whether the figure is 0.1% or -0.1%.

But is all this being too optimistic? In order to only contract by 0.1% there would have to be a very strong performance in the service sector for November and all sectors in December.

One of the more influential reports is the Markit/CIPS purchasing managers' index (PMI).

Its came in at 52.1 in November and 54.0 in December. Any figure above 50 suggests economic expansion.

Chris Williamson, chief economist for Markit, said: "Service sector companies reported surprisingly buoyant business conditions in December."

The manufacturing PMI was 49.6 in November, which was a contraction in the sector, but only a small one, while PMI for the construction sector was positive, showing a figure of 53.2 in December, up from 52.3 in November.

So if the PMI is to be believed, the economy recovered considerably in November and December, after a very difficult October.

The retail sales figures, which are an important component of the service sector, showed a 0.5% fall in November followed by a 0.6% rise in December.

So there does appear to have been improvement after a poor start to the quarter, but will it be enough to mean we see only a small decline in GDP for the quarter as a whole, or have the forecasters been too optimistic about GDP again?

We will find out at 09:30 GMT on Wednesday morning.