My good friend Martin Hewes provided me with a sneak preview of his latest three-year forecast for construction output. It's not a bedtime read for the faint hearted, that's for sure.

The Hewes & Associates numbers show a recession in construction stretching out for the three years of the forecast period, with a nasty 7.4% drop next year. This places it as the most bearish view yet among the industry forecasts.

If this forecast is even near right the industry faces losing 300,000 jobs maybe more.

So how come so bleak?

Firstly some context, it is true that the Hewes forecast does tend to be more bearish than that of the Construction Products Association or Experian Business Strategies, and naturally they do vary sometimes by quite a bit especially at times of great uncertainty, such as now. But none of these forecasts has a tendency towards rash predictions.

The recently released Construction Products Association forecast has construction heading into recession. My view when it was released, indeed the view implicit in the text, was that the risks were heavily weighted on the downside. I suspect some of those downside risks are already being realised.

The Experian forecast, which should be out in a couple of weeks, I expect to be more bearish than the Construction Products Association, but far from as downbeat as Hewes.

But forecasting is a tricky game, particularly when the economy is on the turn and when strident and unpredictable political intervention can turn assumptions on their head. I have long argued that if a forecast is bang on it will most likely be right for the wrong reasons. Also the forecasters are not tested against the real world but against the Government data which present their own set of difficulties.

The real value in forecasts is not in the bald numbers, but in the analysis and assumptions made. And, just as the Construction Products Association did a few weeks ago, Hewes has put together a strong case to back up the figures.

But for me the idea of a three year deep recession in construction comes as a bit of a slap in the face - is it really likely?

The first thing to consider in trying to answer this is what drives construction - and that is four big sectors - private housing, private commercial and the private housing and non-housing RMI sectors.

Folk may tell you all about how much infrastructure work there is, but while it is significant the sector is relatively small. They may say social housing is set for boom times, but it accounts for just 3% of total workload.

In the end the four big sectors mentioned above will determine more or less the shape of the overall construction growth. And it is hard to find comfort great comfort in any of them, given the current economic background.

Hewes has private housing output dropping 22% this year and a further 18% in 2009 before it hits the floor in 2010 when output stabilises. That is a drop of just over a third. Now I have seen plenty of predictions that bleak. And given that urban flat schemes are being hit hardest this may well impact greater on construction output as in general there is more construction work undertaken for each unit produced.

Private commercial - here Hewes sees a significant impacting on the 2009 output figure and stretching through the forecast period. Indeed it is this sector that is largely responsible for extending the forecast recession in construction.

Well, a quick look at the latest monthly orders figures - not that a month's figures mean much - and you can see that the shutters appear to have come down sharpish on new work in that sector. We can expect jitters, but the real question is whether they will turn into full-blown wobbles and workload drying up.

The jury is out on this, but I wouldn't put money on boom times here. This sector is the most sensitive to the vagaries of the global economy. A sniff of a protracted global slowdown may well dampen the ardour of developers - even those looking to build for the next upturn.

Two factors will impact most strongly on private housing RMI, the level of transactions in the housing market and the amount of money in people's pockets. Well, the latest pronouncements by the Bank of England and a host of other experts suggest that the squeeze on cash is likely to persist for some time yet.

Hewes has the biggest fall in this sector coming next year. I suspect his reasoning here is that the second wave effects that will arise from expected increases in unemployment will shape attitudes to spending more next year than this. Who is going to dip into their pocket to have the loft converted when the threat of redundancy is stalking the streets of Britain?

Finally the private non-housing RMI sector. This is a hard one to call as there will be such a mix of factors that determine its performance. But likely to have the most impact will be firms attitudes to cash. We can expect corporate purse strings to be tightening as firms consider how they intend to ride out the current storm - money is in demand and the case for spending it on construction "now" may well prove tough. The most likely outcome is investment budgets will be cut. Also, what impact will company closures have on spending?

Well, yes, looking at the figures, a three-year construction recession is certainly very possible. Indeed it looks increasingly likely. However, my gut feel given what data we have now, is that he depth of the recession in the Hewes forecast looks a shade on the bleak side.

That is far from saying that the forecast will be proven wrong. And, as Martin Hewes regularly tells me, it is better to expect the worst and hope for the best than expect the best and realise the worst.