"Betteridge Law Of Headlines" stipulates that a question in an article headline with a question mark should be answered "no." Usually, the author is not sure about the article thesis, or even suspects it's false, and wants to leave an "out" by questioning his or her own headline. For example, "Could a new miracle drug make you look 20 years younger?" (no!).
In a contrarian spirit, I will break this tradition and will argue that the answer in this case should be a "yes," despite a poor track record of the management overpromising and underdelivering as well as conveniently blaming its predecessors. After all, the payoff, if the management finally keeps its promises, could be tremendous.
The story of last two-and-a-half years of Sterling Construction Company (STRL) can be explained in one simple picture: plotting its steady rising revenue against the collapsing EBITDA. It's clear that something went terribly wrong in 2010 and the company has not fully recovered since.
Revenue vs. EBITDA (TTM) diverged at the end of 2010 (percentage growth last 10 years):
(click to enlarge)
And the equity price has been slammed accordingly.
(click to enlarge)
Sterling is a heavy civil construction company that specializes in the building and reconstruction of transportation and water infrastructure projects in Texas, Utah, Nevada, Arizona, California and other states where there are construction opportunities. Its transportation infrastructure projects include highways, roads, bridges and light rail and its water infrastructure projects include water, wastewater and storm drainage systems. Sterling performs the majority of the work required by its contracts with its own crews and equipment.
In a nutshell, Sterling's two main segments are highways construction and water infrastructure projects. While the company doesn't provide an exact revenue breakdown, it's apparent that most of the revenue comes from highway construction. Unfortunately, the company doesn't provide a lot of specific information about their actual projects such as length, profitability, or allocated resources. Therefore, the investors have to do a lot of guesswork based on the overall reported numbers. Here's an excerpt from its 2012 10K:
"Although we describe our business in this report in terms of the services we provide, our base of customers and the geographic areas in which we operate, we have concluded that our operations comprise one reportable segment and one reporting unit component: heavy civil infrastructure construction. In making this determination, we considered that each project has similar characteristics, includes similar services and similar types of customers and is subject to similar regulatory and economic environments. We organize, evaluate and manage our financial information around each project when making operating decisions and assessing our overall performance."
Almost all the money is provided by either Federal Government, which usually gives grants to the states for construction, or State Governments. There's been much written about poor state of the US infrastructure and, even with constrained government budgets, the infrastructure spending is likely to grow as it enjoys bipartisan support. Just last April, an arch-conservative Texas governor Rick Perry signed a major $2 billion water bill (www.reuters.com/article/2013/05/28/us-us...).
The most important Federal Government spending are MAP-21 and TIFIA, which are fully funded and have been even spared the sequester as most of the money comes from designated gas tax and user fees.
From Sterling presentation:
(click to enlarge)
The three largest customers have been Texas, Utah, and California departments of transportation but their shares are very volatile year-over-year.
Largest customers with over 10% of revenue (2012 10K):
(click to enlarge)
The company also does a significant amount of work in Arizona and Nevada and plans to expand in other Western or Southwestern States.
Current states and expansion opportunities (Sterling investor presentation):
(click to enlarge)
Sterling has grown its service profile and geographic reach both organically and through acquisitions. Expansions into Utah, Arizona and California were achieved with the 2009 acquisition of RLW and the 2011 acquisitions of JBC and Myers, respectively. Unfortunately, these acquisitions have been made at the time Sterling started to struggle.
Is the management credibility compromised?
If we are to believe the Sterling's management today, all problems can be traced to three troubled projects that have had significant cost overruns and inadequate gross margins due to estimation errors. The new projects and the backlog margins are better and the profitability should improve in the 3rd quarter this year.
But before getting into details, let's give a little bit of a background. Sterling has been steadily profitable since early 2000's. Its net income steadily grew all the way through the Great Recession of 2008.
However, after about two year delay, Sterling was hit with two severe consequences of the recession. First, the state capital spending was frozen as most states were still required to balance its books, even with a diminished revenue base. Second, desperate residential and private commercial construction companies which didn't specialize in highway construction, faced with tremendous overcapacity, started bidding on public construction projects at rock bottom prices. Inexplicably, Sterling's management instead of pulling back to protect its margins, decided to continue its acquisition spree and kept on aggressively pursuing the project work.
To make matters worse, its auditor found a "material weakness" in 2011 and 2012 financial statements.
"In the fourth quarter of 2011, management identified a material weakness related to the established process for estimating revenues and costs on its construction projects. The accuracy of our revenue and profit recognition in a given period is dependent on the accuracy of our estimates of the revenues and costs to finish uncompleted contracts."
While this is not as bad as "material misstatement," which forces the company to restate its filings, it may still mean that some of the numbers may not have been accurate in 2011 and 2012.
According to PACOB (Public Company Accounting Oversight Board):
"A material weakness is a deficiency, or a combination of deficiencies, in internal control over financial reporting, such that there is a reasonable possibility that a material misstatement of the company's annual or interim financial statements will not be prevented or detected on a timely basis."
It's likely that a "material weakness" is closely related to the underbidding on the projects as it mentioned errors in estimating the revenues and gross margins.
Last two years have been outright ugly for the company and its shareholders. Sterling experienced heavy losses due to write down on three unnamed (unfortunately!) projects; two in Texas and one in Arizona, while the overall gross margins have compressed to low single digits.
(click to enlarge)
The current management lays the blame for last year and this quarter losses squarely on the shoulders of its predecessors. The new CEO, Peter E. MacKenna, who came on board in September 2012, claims that the losses are due to low margin contracts signed just before his tenure. MacKenna had a long tenure as CEO of Skanska AB, a Fortune 500 public company and one of the ten largest construction companies in the world, so his experience is not in question. He replaced a long term CEO Patrick Manning (1998-2012), who still remained an executive and the Chairman of the Board of Directors. It's reasonable to assume that there must be at least some truth in MacKenna's statements as his predecessor is still boss as the chairman of the board of directors. Nevertheless, the investors are always skeptical when the current management blames its predecessors.
Even if we take MacKenna's words for their face value, he already overpromised and underdelivered on Q3 2013 conference call.
On Q2 conference call (08/09/2013), he promised the troubled projects will be wound down by mid-2014:
The declines in gross margins were primarily attributable to 3 construction projects, 2 in Texas and 1 in Arizona, which is being constructed by our Utah subsidiary.
The average margin on these earlier awards continue to be impacted by the Q2 gross profit write-downs and is now estimated currently at around 2%. We expect to complete work on a significant portion of these low-margin contracts in the next 9 to 12 months.
In Q3 2013 conference call, he showed much more optimism (11/11/2013):
Approximately, 17% or $118 million of our backlog is attributable to awards prior to 2012 and carries an average gross margin of 1.8%. This backlog is scheduled to be substantially completed in early 2014.
To the profit warning issued (02/04/2014), which again lowered the expectations:
The Company attributed the losses primarily to additional write-downs on three large projects booked prior to 2012 in Texas. These projects, which are expected to be substantially complete by the end of the second quarter of 2014, have been encountering unanticipated net revisions to contract cost to complete estimates that became apparent and quantifiable during the 2013 fourth quarter.
The silver lining in these warnings is that the last update was given with only 4 months left until the end of the second quarter. Unless we believe the management is intentionally misleading, it should have a much better visibility when the trouble contracts will be finished.
Sterling had remarkably stable gross profit margin around 10% until 2011. It apparently did not bid on the low margin projects until 2010.
(click to enlarge)
There are three basic types of bids on the contracts (with several variations): fixed bid, cost plus, and engineering consulting. Sterling's contracts are almost all fixed bid, where the bidder assumes all risks of cost overruns. The contracts are usually awarded to the lowest bidder, who also met qualifications requirements. In last few years, the lowest bid by itself is no guarantee of getting the contracts due to intensified competition, as the clients sometimes do an "extra round" of bidding to further squeeze the costs. Recently Sterling revised its backlog recognition, limiting it to only those projects that have been awarded but excluding the projects where they placed the lowest bid but the award wasn't final. I have every reason to think that Sterling backlog is understating its future revenue, compared to the earlier years where it included all lowest bids.
The backlog continued to grow uninterrupted through the years and, in general, has been a good predictor of the revenue (although it dipped together with the revenue a bit last quarter due to a completion of large projects in Utah). The gross margins have become very volatile in 2010. Because of a material weakness identified by Grant Thornton, the firm's auditor, 2011 and 2012 filings may have inaccurate estimates for gross margins and backlog. In the same period, Sterling made several acquisitions and took several write-offs making financial data very "noisy." It's likely that we cannot fully rely on 2010-2012 gross margin estimates and should look at 2006-2009 period instead to understand the historical trends.
Backlog vs. revenue (author's calculations):
(click to enlarge)
It's also worth noting that in 2012, Sterling adopted a more conservative backlog estimate where it counts only awarded projects. Therefore, it's expected that book-to-bill ratio would become somewhat lower than its historical average. According to Sterling's 2012 10K, the backlog is usually realized in 12-18 month period. Any unfinished work can be canceled by a client any time, which, nevertheless, is a rare occurrence according to Sterling.
Total backlog as of September 30, was $694 million, compared to $714 million at the end of the second quarter. Current backlog contains $261 million of projects awarded in 2012 at an average margin of 6.3% and $314 million of projects awarded in 2013 at an average gross margin of 8.3%.
While the gross margin has been historically low last year, the backlog gross margin is edging up to its historical 10%+ average. If Sterling continues to avoid low-margin projects after finishing the three troubled projects in Q2 2014, we probably should expect the gross margin above 8% for the second half of the year.
With the SG&A expense percentage relatively stable and the future revenue based on its backlog, we can build a "pro-forma" statement of what Sterling can earn when (or if) it will put its troubled projects behind it.
Revenue and earnings projections (author's calculations):
Note that pro-forma 2014 only covers last 6 months of 2014.
I made the following assumptions: the gross margin in the second half of 2014 will match the expected gross margin of the reported backlog of the projects awarded in 2013. In 2015, the gross margin will match the historical average before the "Great Depression." SG&A margin will be the average of last 3 years (after all major acquisitions). Revenue will grow at half of historical average of the growth of the backlog (5.9%). There is some evidence that at least some contacts from the backlog was never actually fulfilled based on the historical observation of the book-to-bill ratio exceeding the forward growth rate. I use the base year 2012 as the revenue offset as 2013 had a dip of revenue because of the timing of large Utah contracts. CapEx and Depreciation and Amortization will grow proportional to the revenue growth.
Mathew Paul of Sidoti, one of the few sell side analysts who still follow Sterling, projected similar revenue numbers. He believes Sterling will have a $648 million revenue in 2014 and $713 million in 2015. He thinks the company will generate $0.94 EPS in 2015 and $10.6 million of free cash flow. In short, his projected revenue is higher than my estimate but the margins and earnings are quite a bit lower.
I believe the difference in earning vs. revenue may arise from additional losses Sterling experienced after the analyst report. We'll need to wait for the latest 10K to see the actual size of Net Operating Losses benefit but we can deduce from Q3 10Q that Sterling derived nearly $13 million tax benefit in 2013.
Tax expense and benefit in year 2012 and 2013 (Q3 2013 10Q):
(click to enlarge)
With losses projected to be around $47 million this year, it's likely that Sterling will pay no taxes in 2013 and very little in 2015 (not all losses are tax deductible). With little net debt and no taxes, most of the (EBITDA-CapEx) will flow to the bottom line. Therefore, Sterling may be able to generate as much as $40 million net income and free cash flow in 2014 or about $2.38 per share.
I must caution investors that earning $2.38 in 2015 is the best case scenario. It assumes that the management is accurate in its estimates of the future gross margins and the three troubled projects will be done by the middle of this year. It also assumes a modest organic revenue growth based on the past performance. So, before you apply a 15x multiple to 2015 projected earnings to fancy that you are buying $35 stock for under $10, consider the risks. Just three weeks ago, Sterling issued a profit warning, pushing the finish of the unprofitable projects from the beginning of the year to the end of the second quarter. The company also mentioned violating debt covenants but assured that the resolution is in place (probably not a source of a major concern since the company has net cash on its balance sheet).
All in all, if everything goes well, you may be looking at 200-300% return in a year but if not, you will book a moderate loss.