How far will construction costs fall?

Friday, 6 March 2009

Here is an excellent research paper from Turner & Townsend, quantity surveyors.

How far will construction costs fall? Five percent? Ten percent? What about thirty percent? We are already hearing stories of falls of as much as 15% in construction costs in some areas and on some jobs. However this has not been our observation so far. This week we assess how far construction costs are really likely to fall.
 
As we all know, construction costs in Australia have transitioned from a long period of growth, (quite high in some regions) to a period where construction costs are falling. As a result we have adjusted the cost data in the subscription area of CostWeb to reflect this.
 
Our assessment of falling construction costs is being tempered at this stage. Most of the evidence so far is anecdotal or based on a few trade rate drops. Based on the amount of bad news it would be tempting to forecast substantial falls in construction costs. We prefer to look at the evidence and the economics. So here goes.
 
Substantial falls in formwork, reinforcement and structural steel prices are being observed. The former are because some new jobs are being cancelled, and the latter are because the world price of scrap steel has fallen. Copper has also fallen and this will help ease electrical cable and copper piping costs.
 
Whilst retail prices for concrete are still showing publicised price increases, some commercial concrete supply prices after negotiation appear to be falling by as much as 5%. Low delivery costs will give some further relief but at this stage 10% or 15% falls appear unlikely.
 
Certainly we are hearing that many previously busy subcontractors are now wondering where their next job will come from, and a big increase in enquiries and subcontractor bids are coming in. But does that mean overall construction costs will drop substantially? After all, for construction costs to drop by say 15% every trade needs to fall on average by 15%. Until we see how badly (or well) the economy fares we are reluctant to forecast large drops (e.g. 15%) in construction costs.
 
Construction costs in most areas are expected to be either flat or fall modestly in 2009 and 2010 before starting to increase in line with an economic recovery during 2011. Much depends on the type of job. If the job is straightforward, has good build-ability, requires lower levels of trade skills, and has minimal green-star aspirations, then there is more scope for costs to fall perhaps by 10% or more under a competitive tender scenario. For most other building types we expect costs to fall by up to 5% depending on job and region.
 
What about the evidence of past downturns?
 
Yes but what if the economy really deteriorates and unemployment goes well beyond the governments forecast of 7%? In other words the trade skills shortages we have been experiencing are well and truly over and subcontractors become really desperate.
 
That was the case during 1991-93 (the “recession we had to have”) when unemployment increased from 5.9% to 10.7% over a period of 3 years.
 
At the same time there were large falls in some major trades. Formwork fell by up to 30%, electrical trades fell by 18%, structural steel, plumbing and blockwork fell by 15%.
 
Yet look at the attached chart which shows how much construction costs increased in the period leading up to the recession we had to have and how much construction costs then fell. Looking region by region you will see that the states which fell the most also came off several years of overheated construction cost growth.

Over two years construction costs in Melbourne fell by 14% and in Sydney by 9%. Most other cities saw only modest falls or no falls at all. In other words, the bad news built into some large trade falls did not generate as big a fall in the Building Price Indexes as might be expected. In the boom years prior to the recession Melbourne, Sydney and Perth had experienced strong construction cost growth, so there was further for costs to correct  downwards. Notice that costs in Perth did not fall at all, but experienced a long period of stagnation.
 
When we enter our trade rate observations and forecasts into CostWeb models we see either modest falls in overall construction costs or flat construction costs for the next two years.
 
Why is this?
 
Building material prices are unlikely to see further big falls.
 
So far most building material prices trades are holding up, as a result of strong existing demand for materials. The evidence points to that strong demand continuing. The December data for private investment expenditure (both actual and forecast) was surprisingly strong. According to the ABS survey of 8000 businesses conducted in December, businesses investment intentions for 2009 have actually increased since a year ago. There was also a 6% increase in actual business investment in December over the previous quarter. This indicates strong investment momentum going into 2009.
 
Our low exchange rate is also keeping the prices of imported building materials up. A sustained weaker dollar will inevitably hold up the prices of whitegoods, vehicles, electronics, machine tools, granites and marbles, lifts and escalators, sanitary ware and plumbing fixtures.
 
Those building material prices which have fallen are mostly those items which have been exposed to the commodity price cycle which is now over. Commodity prices (copper, aluminium, zinc, nickel) are back to 2005 levels and unlikely to fall much further.  These same materials are also hostage to the exchange rate weakness and it is reasonable to assume that further falls in landed price are not on the cards.
 
Wage rates are holding up.
 
As well as materials, wage rates are likely to hold up. The Treasury is still forecasting that wages will increase by as much as 3.25% in 2009-2010. Recent surveys by the Housing Industry Authority indicate that trade skill shortages still persist. The enterprise bargaining system and the skills shortages situation are two factors that will prevent wages from falling.
 
So if building material prices are likely to hold up and wages hold up where will large falls in construction costs come from? The obvious candidate is a fall in margin at the head contractor level and the subcontractor level.
 
But a large fall in contractors’ margin is unlikely too.
 
Turning to head contractors first. In cities such as Perth and Brisbane margins were very low in the late 1990s because there was so much competition in the market. Could we return to similar low margins? We think not, although of course margins will be reduced a little. Since the late 1990s, many current building contractors have become very different entities. Most are now publicly listed companies where the shareholders would not tolerate drastically reduced profits. Modern contractors tend to specialise and perform design and construct contracts, in contrast to the late 90’s when competitive tenders were let after the design had been finalised and drawn up. Today’s form of tendering provides more opportunity for head contractors to seek win-win opportunities that preserve their margin.
 
Many sub-contractors are specialised too and far more technically and financially sophisticated than in previous decades. For some trades there are only a few subcontractors with the capability to perform the work. Hence the opportunity to squeeze subcontractor margins is also substantially less then it was in the 1990s.
 
Most head contractors are very wary of big falls in subcontractor prices, especially where there is a perception that the subcontractor is desperate for the work to survive. Invariably reputation and financial security take as much relevance as price in choosing subcontractors. 
 
So what is the case for a large fall in construction costs?
 
Surely a large increase in unemployment will force costs to fall steeply?
 
The Treasury is forecasting unemployment to increase to 7% from the current level of 4.8%. This is still a low level of unemployment and unlikely to provide the trade skill surpluses necessary for wholesale falls in construction cost?
 
Yes, but what if the Treasury is wrong?  What if the unemployment follows the 1991-93 route and blows out to 11%.
 
Our view is that unemployment is unlikely to blow out to the same extent as 1993.
 
The circumstances were different in 1989-1993. Businesses in general were more highly leveraged then than now. When interest rates peaked at 17% many could not afford the interest and there were numerous bankruptcies. For this reason we are still more comfortable with the Treasury assessment that unemployment will peak at around 7%.
 
We have already discussed the high level of private business investment intentions in the ABS data. Some of this is already committed to construction jobs underway and will help keep construction momentum going through 2009. On top of this we need to add the high level of government public works, and infrastructure which is also underway and growing. 
 
Next, the stimulus package and its emphasis on shovel ready projects in schools public housing and insulation will provide a countercyclical boost to the economy by the end of 2009, and help to keep up the demand for the smaller subcontractor businesses that do not operate in the engineering, infrastructure or larger public works market.
 
Finally the low interest rate environment will spark a modest revival in residential housing construction during late 2009, which will grow steadily through 2010 as the economy recovers, and confidence and job security returns. Already affordability has improved dramatically with low interest rates, and the rent/buy equation is swinging in favour of buying, especially if the First Home Owners Grant is available.
 
Finance approvals are indicating the start of a residential upturn and despite poor January data we should see improved residential building approvals follow suit as the effect of low interest rates becomes apparent. Low end house building should recover from now.  
 
For these reasons we think that falls in construction costs are likely to be modest rather than cataclysmic. Construction costs are likely to drop modestly and stay flat for several years as the cycle corrects back to the long term trend.
 
Our arguments are summarised in the attached table.

By Gary Emmett, Turner & Townsend