Construction businesses are always difficult to understand and value. The markets in which they operate are varied and volatile, and construction markets worldwide are transforming as they embrace modern construction technologies and methods.

These markets are now being challenged to respond to the global industrialisation of construction and the sourcing of its pieces and parts from multiple jurisdictions. According to Turner & Townsend’s 2015 International Construction Market Survey, the typical construction margin on a medium-sized construction project in Sydney and Melbourne is reported to be four per cent. In London, Singapore and New York, those margins are estimated to range between five and six per cent. Sydney and Melbourne were ranked the fifth and 12th most expensive places respectively to build prestige high rise CBD offices out of 35 global cities the report studied.

Lend Lease is Australia’s premier home grown integrated property development and construction enterprise, and its 2015/16 Annual Report has just been published. This year, the company’s gross revenue was $15.105 billion, up by 14 per cent from $13.299 billion last year. Lend Lease’s earnings before interest and tax was $1.054 million (EBITDA), up nine per cent from $0.97 million for the same period last year. That’s a group margin of around seven per cent on the gross revenue for 2015/16.

The construction operations made up nearly 80 per cent ($12.03 billion) of these revenues. The group’s construction turnover grew by $1.095 billion, or nearly 10 per cent, while the Australian construction turnover grew by $358 million or 6.1 per cent. In Australia, the construction business contributed $6.3 billion to group revenue and delivered earnings before interest and tax of $232 million or around 3.7 per cent on turnover. The group wide construction margins averaged 2.4 per cent, down from 2.6 per cent the prior year.

On the face of the annual report, Lend Lease’s construction arm enjoys a captive development client. But the quantity of earnings may be overshadowing their quality.

Another way of communicating the group’s performance was reported by the Australian Financial Review in August.

“The global (Lend Lease) property giant benefited from the profits on housing, and from new office towers, to boost net profit in 2016 by 13 percent, to $698.1 million,” the AFR stated.

The AFR reported that the development business’ forward pipeline rose by an additional nine per cent to $48.4 billion. In construction, the backlog rose by 20 per cent to $20.7 billion, with another $7 billion at preferred bid stage. The publication added that the construction business operating EBITDA rose just three per cent, with a strong result in Australia eroded by weaker margins elsewhere, particularly as the business mix changed in the US. This report in my view is essentially about the quantity of earnings.

Lend Lease’s 2016 Report shows that in a very buoyant property development market, the unit’s revenue increased from $1.875 billion to $2.544 billion for the year, with earnings before interest and tax of $500 million or a margin of about 19.7 per cent. The Australian development operations contributed a margin of $392 million on its revenue of $2.034 billion or about 19.3 per cent, down from a return of 22.3 per cent for the year prior. On the face of these results, it may be arguable that the construction operation is having a dilutive effect on the group.

The construction backlog includes $3.9 billion in engineering projects and $1.7 billion in services contracts. Lend Lease purchased the Valemus businesses comprising the former Baulderstone, Abigroup and Conneq businesses in late 2010. This acquisition cost Lend Lease about $1.04 million, and at the time was reported to be adding an estimated $5 billion in forward orders to increase the groups work book to $13 billion. Data from the sale of Valemus suggested that the earnings for the coming year were estimated to be $116 million on a revenue of $4.6 billion or around 2.5 per cent. It’s interesting that in the 2015/16 Lend Lease annual report, the present engineering and services revenue forecast are exactly $4.6 billion. There is no information on the make-up of expected earnings for this segment of the business.

Lend Lease is committed to leadership in the sustainability of its projects. A recent initiative has been to explore the potential of timber buildings. During 2015/16, the company has made a number of announcements and commissioned articles about how timber construction can be faster and cheaper. The company’s third local timber building is now under construction at Barangaroo. There are no details of this new direction in the annual report or of any strategic intent as to how this new construction methodology may be deployed to improve the future quality of earnings. It would be a pity if this initiative was not about measurably faster and cheaper construction in addition to green credentialing.

Included in the Lend Lease earnings for 2015/16 was the profit attributed to the sale of assets as reported to the ASX on June 2 of this year. In the notes attached to the accounts, a contribution of $151.6 million was made from the sale of interests in the Sunshine Coast University Hospital, the ICC Darling Harbor, Sydney Towers Trust and One International Towers in Australia, and the European operation’s Stafford City Business district.

These transactions contributed 14.4 percent to the group’s total earnings of $1054.9 million. Coincidentally, this contribution is about equal to the reported growth in earnings this year. In my mind, this earnings enhancement is reminiscent of the transactions that used to occur when GPT was a captive of Lend Lease and occasionally where expensive assets were warehoused for another day. It is perhaps fortunate that the global hunt for low positive yields can make even expensive assets attractive today. The Gold Coast University Hospital is another infrastructure project undertaken by Lend Lease. This project was reported as the seventh most expensive project in the world when it was under construction.

The Lend Lease accounts explain how the executive remuneration scheme works for the eight top executives who have a combined actual remuneration of $29.3 million in 2015/16. The increase in total compensation and incentives appears to have risen on average by 6.7 per cent. Lend Lease builds its incentive scheme around five pillars of value. All are worthy, but the driver of the financial pillar would seem to centre around the quantity of earnings and their influence on the company’s share price over the last five years. The impact of measures addressing the quality of earnings is not obvious (such as details of earnings related to the construction company’s performance comparing estimated or bid prices, with actual cost.)

Lend Lease asserts that its competitive advantage is an ability to leverage a vertically integrated business model and financial strength to undertake large, complex, long-term developments deploying end-to-end property infrastructure solutions. The company points to a long dated portfolio of infrastructure projects as a core strength. On this basis, Lend Lease’s ability to leverage value creation through its vertically integrated business model should be demonstrable by benchmarking these with peers.

One might reasonably expect that the construction business should be capable of delivering higher quality earning than achieved, given that it is the beneficiary of a vertically integrated end to end business model. In my view, these should outperform those of a downstream moderate sized construction company. Investors may then expect the material factors that affect these performances to be spotted and turned around. The alternative is to question the business unit.

The Lend Lease construction arm appears to be an expensive operation. It is amongst the first to sign up to new union Enterprise Bargaining Agreements that soon flow out across the industry without any obvious quid pro quo to lift the company’s productivity. My opinion of Lend Lease construction projects based on my own observations is that productivity is sub-optimal. And, in my view this relates as much to inadequate construction planning before work commences on site as it does once it commences. The utilization of cranes, temporary construction infrastructure and the site workforce seems to me to be likely to erode margins. I cannot believe Lend Lease’s executives would sanction the pursuit or acceptance of projects with starting construction margins like those recently reported.

I believe it is time for Lend Lease to do a roots and branches analysis of the in-house construction business. There are, in my view, equally able, more efficient and innovative construction alternates who could competitively undertake projects for an informed development and procurement client who focuses on extracting the best through-the-cycle returns for investors. Here the quality and quantity of earnings should be of equal standing.

Mirvac has just released its 2015/16 Annual Report. I believe the discipline and clarity that Mirvac’s CEO Susan Lloyd-Hurwitz has brought to this business over the last few years could point the way for mixed property and construction businesses. Mirvac now emphasizes the soundness of its developed, held and managed investments in offices, industrial and retail portfolios to underpin both the quality and quantity of the business’s earnings going forward. While the residential development play in Mirvac remains important, Lloyd-Hurwitz now has the ability to manage a strategically smaller part of this company’s balance sheet than the highly cycle exposed and undisciplined portfolio she inherited.

Mirvac seems as enthusiastic about large urban renewal projects as Lend Lease. These combine most of the vertically integrated elements of the Lend Lease business with the option of better managing future exposure to construction industry vagaries and perhaps a fat in-house construction business. This is not to say Mirvac is unsentimental about self-performing projects in-house, but it has this as an option that does not equate to the nearly 80 per cent of turnover the Lend Lease construction business does, until it recalibrates.

And finally this conversation raises the value-for-money question for clients who may feel that any price can be justified to get a blue chip to take on their construction risk. This scenario is not dissimilar to the one where in the 1970s and 80s it was often joked in large organisations that ‘no one ever got sacked by awarding a contract to IBM.’

Lend Lease performs many Infrastructure PPPs and wrapped investment projects. In the end, there is nothing more expensive than carrying into the future a saddle of debt that comes from unchecked initial construction costs. And there is nothing more damaging to the economy when unchecked costs can be simply passed on when there is little or no resistance.