Description

Summary investment thesis

Following a near doubling of the stock in the last 2 months, also fuelled by a short squeeze, we believe TPC represents a very attractive opportunity to short a structural cash burner with potential 100% downside in a bear case scenario and very little upside in a base case scenario. We believe the short squeeze was triggered by a very superficial reading of the company’s latest set of quarterly results. In Q3-19, TPC reported their best ever quarterly free flow, equal to $200m. As we will explain below, we believe this seemingly positive development is nothing but a smokescreen to cover up deep fundamental issues with this business. We think there is a high likelihood TPC will need to resort to a rescue capital increase to salvage the situation, leading to substantial downside to current share price.

Summary TPC history and business description

TPC today is one of the largest contractors in the US, ranked 10th overall:

TPC has a long and interesting history. Perini Corp. has been providing construction services for over 100 years. In 1999, Ronald Tutor, the CEO and main shareholder of Tutor-Saliba, a California based contractor, became the Chairman and CEO of Perini. Perini Corp. ended up acquiring Tutor-Saliba via a merger in 2008, creating what is now Tutor Perini (“TPC”). Ron Tutor, going for 80, is still the CEO of the company. TPC core business was general construction and building services. In particular, TPC targeted 2 key markets: large building projects in the leisure / gambling / health sectors, especially in Nevada (think  of Caesars Palace and MGM) and some civil projects in the Northeast, primarily in Boston and NYC. The former was the core business of Perini, the latter was the business of Ron Tutor, Tutor-Saliba.

Under the stewardship of Ron Tutor, the business grew dramatically, especially in its core building segment. In the first decade since Ron Tutor became CEO, TPC sales grew from little over $600m in 1999 to over $5bn by 2009. The building segment alone grew to over $5bn in sales by 2008. Growth was fuelled primarily by large projects in Las Vegas where TPC had a stronghold. TPC had a strong reputation for efficient operator. In 2005 Forbes named Perini (now TPC) “One of the Best Managed Companies in America,” ranking it No. 1 in the construction industry. By 2008, this business was on fire, the Hospitality and Gaming segment alone contributed $3.7bn in sales. TPC was involved in the construction services of the MGM, MIRAGE CityCenter, Cosmopolitan Resort and Casino, Wynn Encore Hotel and Planet Hollywood’s Westgate Tower. In 2008, the building sector alone contributed $152m in EBIT, the highest ever recorded for TPC in that segment. TPC was the established leader in Building entertainment:

Things started to deteriorate quickly though. The GFC hit and large projects in Vegas were abandoned. The building segment imploded with sales going from $5.1bn in 2008 to $1.5bn by 2012, with obvious impact on margins:

To compensate for losses in the Building segment, Ron Tutor, right after Perini merged with his own Tutor-Saliba company, embarked on an acquisition spree, focusing on the segment he was familiar with: Civil. Ron Tutor acquired several businesses in the Civil space, such as Lunda Construction, Becho and Frontier-Kempter in the Civil space and Fisk Electric and Five Star Electric in the Specialty Contractor space. These transactions helped the Civil segment and the Specialty Contractor segment grow sales between from $0.5bn in 2008 to $2.6bn by 2012. In our view, this is where TPC problems began. TPC and his ambitious CEO simply overstretched themselves.

It’s important to note that the Civil segment is a much trickier one than building. Projects are typically longer and more complicated. Cost overruns are common and for this reason, typical contractors try to limit their exposure by having variable pricing or cost-plus type of contracts, rather than fixed price contracts. Furthermore, large Civil projects are often subject to many variation orders during the construction period, leading to tricky contractual resolutions. Finally, while the building segment is predominantly dealing with private clients, the Civil segment is predominantly dealing with public entities, local municipalities or the federal government. Competition is somewhat lower as bidders require scale and expertise. Because of this, margins could in theory be better than in Building, were contractor earn LSD margins. Other civil competitors such as Fluor, KBR, Parsons, Granite, Skanska, Dragados, Kiewit Corporation etc. can earn mid-to-high single margins on larger project if managed well. Tutor Perini is still a smaller competitor, ranked 10th in the US but 2nd overall in the transportation (roads, bridges, tunnels) segment. It is therefore extremely puzzling to see TPC generating double digit EBIT margins in its Civil segment when all peers make mid-to-low single digit margins.

The Civil segment has been the real driver of performance for TPC in recent years, representing now c. ¾ of the EBIT generated by the company before central costs and over half of its backlog. Civil revenues went from less than $400m in 2009 to nearly $2bn in 2015. Run Tutor was absolutely determined to have the business grow back to the heyday of the Las Vegas boom through aggressive growth in Civil. He nearly succeeded as by 2016, TPC sales were back to $5bn from little over $3bn in 2010.

The growth in Civil though came at significant risks for TPC. Large construction projects are complicated. Conservative players will try to minimise risks structuring contracts where excess costs will be borne by the customer, or at least shared. Not so Tutor Perini – the proportion of contracts under fixed prices increased dramatically over time and it represented 80% of sales in 2018:

Peers such as Granite, Fluor, Skanska typically keep this percentage between 30% and 60%. TPC has effectively gone for broke, betting the house on potentially lucrative contracts that could become huge liabilities if mismanaged. Think about a $4bn contract where TPC aims to make 10% margin. A 20% cost overrun will crush margins to -10%. This is exactly what TPC has been doing since 2009.

Aggressive Accounting issues

The problem with construction companies is that revenue recognition is subject to management own estimates. “Because control transfers over time, revenue is recognized to the extent of progress towards completion of the performance obligations. The selection of the method to measure progress towards completion requires judgment and is based on the nature of the products or services provided… Due to the nature of the work required to be performed on many of the Company’s performance obligations, estimating total revenue and cost at completion is complex, subject to many variables and requires significant judgment”. This significant issue is compounded by the fact that larger Civil projects are naturally subject to order variations and changes to the original contract. The contractor continues to incur in costs and recognises revenues and margins associated with these costs, not knowing know though whether the client will ever pay for these.

TPC clearly highlights this problem amongst the risk factors in the annual report:

• “If we are unable to accurately estimate contract risks, revenue or costs, the timing of new awards, or the pace of project execution, we may incur a loss or achieve lower than anticipated profit”
• “Our contracts require us to perform extra, or change order, work which can result in disputes or claims and adversely affect our working capital, profits and cash flows”
• “Our actual results could differ from the assumptions and estimates used to prepare our financial statements.”

Put it very simply, TPC highlights 3 main risks with this business:

• TPC may sign contracts that are loss making
• Because these contracts are loss making, TPC will end up burning cash
• TPC may recognise profits on such contracts when in reality they are loss making

This is the crux of our thesis: TPC has signed loss making contracts since 2010, burning cash throughout, while telling the market they have been generating healthy margins.

This problem expresses itself at the balance sheet level under “Costs and estimated earnings in excess of billings”, otherwise known as Unbilled Receivables. These represent the receivables that the client owns the contractor for revenue recognised by TPC where costs have been incurred on a cash level but the client couldn’t be billed. The 2 larger categories of unbilled receivables are “claims” and “unapproved change orders”. Claims occur when there is a dispute regarding both a change in the scope of work and the price associated with that change. Unapproved change orders occur when a change in the scope of work results in additional work being performed before the parties have agreed on the corresponding change in the contract price. Over the last 9 years, since TPC focused its prospects on the Civil segment, unbilled receivables grew exponentially from little over $100m in 2010 to nearly $1.2bn in 2019.

Particularly worrying was the growth in claims, growing from close to zero in 2010 to over $700m in 2019:

Claims are the most troublesome item on balance sheet because it represents revenues (and margins) that have been recognised in the P&L that are highly uncertain in terms of collectability.  The clients dispute both the change in the scope (“I never told you to add the extra ditch to the foundation of that bridge...”) and the price associated with that change (“you cannot possibly charge me $1m for that extra ditch”). Over the last 9 years, TPC recognised c. $1bn in revenue that not only wasn’t paid for by the customer, it wasn’t even billed.

The issue becomes apparent when we analyse cash flow. The gap between reported EBIT and free cash flow since 2009 is striking:

The change in fortunes at TPC since the company decided to aggressively move into the Civil segment is remarkable. Between 2003 and 2010, when TPC was involved in Las Vegas type of construction projects, they generated c. $500m in free cash flow, which is over 50% of the EBIT generated in the period notwithstanding the negative FCF reported during the GFC in 2009. Vice versa, between 2011 and 2019, TPC generated c. $1.6bn in cumulative EBIT and negative cumulative free cash flow. Over the last decade, TPC did not generate a single dollar in cumulative free cash flow.

The above suggests that TPC continued to recognise P&L profits on contracts that were essentially unprofitable. The huge balance of claims will never be entirely paid off as some of these receivables are not going to be recognised by clients.

Several large impairments made in recent years suggest the above interpretation of financials is correct. Following the numerous acquisitions discussed above, TPC had nearly $900m of Goodwill on balance sheet by 2011. Over the years, the company was forced to take several impairments to these acquisitions, the latest in Q2-19 in the Civil sector. Following these impairments, Goodwill today is only $205m. There have been in the last years impairments totalling nearly $700m suggesting the true underlying value of the acquired business is significantly lower than previously expected. Still, TPC didn’t write off any receivable nor restated previously recognised revenue. We believe the day of reckoning is approaching fast.

Stretched balance sheet

The inability to generate cash discussed above, led the company to incur in substantial amounts of debt to fund operations. Large contractors should ideally be debt free, given the capital-intensive nature of their business. Because TPC couldn’t generate any cash, net debt went from $450m in 2011 to nearly $800m, bringing total leverage to very uncomfortable levels.

We actually believe the situation is much worse than it looks. The company obviously reports only debt balance at quarter end. Intra-quarter, we think the company is continuously drawing on its revolver to fund day to day operations. This shouldn’t happen at a well-run, large scale contractor. The average cost of debt for TPC should be c. 6%, give or take:

However, if we calculate the actual interest cost on a quarterly basis, we notice that the implied interest rate paid by TPC on the average debt balance of the period is much higher than this:

The only way to explain this is to assume heavy intra-quarter revolver borrowings. Doing a simple back of the envelope calculations, we can imply from the above that on average, TPC carries at least $300m of debt on balance sheet more than it reports at quarters end. Average cost of debt is c. 200bps higher than it should be, or c. $15m a year. The revolver costs c. 4% a year. 4% of $300m is $12m. Tutor Perini is financially much more stressed than it would like you to believe.

Chronic inability to collect

Since 2010, when Ron Tutor embarked on an aggressive growth strategy in the Civil segment, TPC failed to collect unbilled receivables. Put it differently, TPC continued to generate positive EBIT on paper but negative free cash flow. Over the years, this became an important focus for management for 2 reasons:

1. Notwithstanding record backlog and good growth in earnings, the stock wasn’t going anywhere as investors realised that TPC was unable to collect on its unbilled receivables
2. TPC was starved for cash and needed cash quickly. Collection of old receivables would be an obvious place to start in order to generate cash

What we try to illustrate below is management chronic inability to collect receivables notwithstanding their bombastic claims to the contrary. Ron Tutor promised time and time again he’ll tackle the issue, but his track record should really make investors very sceptic.

Unbilled receivables increased from $139m in Q4-10 to $905m in Q4-15. At Q4-15 results, in February 2016, Ronald Tutor finally had to concede to the market that the company had an issue with collections and that the unbilled receivable balance needed to be resolved. As far as I can tell, it’s the first time he tackled the issue publicly. This is what he said:

“The increase in the cost in excess account has certainly gotten our attention and we are working hard to drive it down. Note that just three years ago the cost in excess balance was about half of what it currently is. We expect to reach that level by no later than the fourth quarter of 2017”. 

Quarter by quarter, management kept repeating this mantra. Just 3 months later, at Q1-16 results, the CFO said “in late February, we communicated our plan to reduce unbilled costs, that's the cost and estimated earnings in excess of billings reflected on our balance sheet from $905 million at the end of 2015, to about half that amount by the end of 2017, so over a two-year period. To accomplish this, we have been intensely focused on resolving numerous claims and unapproved change orders, as well as billing and collecting other unbilled amounts…We're starting to see considerable traction in our efforts as we have made very good progress in the first quarter of 2016".

A quarter later (Q2-16), same promises were reiterated: “We expect to make further progress in reducing our debt level as we collect substantial cash that we are owed throughout the balance of this year and beyond in accordance with our previously stated goal”.

Even as the company was clearly not delivering on its stated plan, it continued to boast confidence in these targets. A quarter later (Q3-16) the CEO stated on the call ” ...this is Ron Tutor. I’m more confident today than the last time we spoke that we will collect the dollars, and we are currently making great strides…”. I won’t bore you with comments from every single quarter, suffice to say that the company missed its targets miserably. 

Unbilled receivables balance was $905m in 2015. The CEO vowed to bring it back to 2012 levels ($465m) by Q4-2017. Instead, unbilled receivables increased a further $28m by Q4-17 to $933m, missing management target by nearly half a billion dollars. 

Having missed its 2017 targets, management then focused on 2018 and 2019. On its Q4-17 call, the CFO stated the following: “we remain dedicated to significantly reducing our unbilled costs. Based on how negotiations are developing and expected to further progress, we anticipate more progress in resolving and reducing certain of our larger unbilled cost issues in 2018 and beyond…our operating cash flow in 2018 is expected to exceed net income…Deleveraging, in other words, debt reduction, remains a top priority in terms of capital allocation. We plan to utilize as much operating cash we generate in 2018 as possible for this purpose”. The annual guidance given at Q4-17 results implied net income at the mid-point of c. $100m. Operating cash flow should have been higher. It ended being $21m, some 80% lower. Furthermore, TPC was supposed to pay down debt during 2018. Instead, net debt increased from $543m to $645m. Once again, management failed to deliver on its promises.

We illustrated the above to give some background to Q3-19 results, where management showed positive cash generation in the quarter and made new promises about cash collections in 2020. Given management track record, we urge investors to treat their promises with extreme caution.

Enter Q3-19 results

Expectations were high going into the quarter. During TPC’s Q2-19 call, the CFO promised strong cash generation in H2: “We expect more substantial reductions in unbilled during the second half of this year and into 2020, as we continue to focus our efforts on negotiating, litigating and settling the various claims”. The stock rallied over 50% from Q2 ahead of Q3, in expectation of a cash flow relief. Note that the stock is heavily shorted (c. 23% S.I.) and forensic accountant specialists like Gradient have a well-known negative thesis on the name, so modest positive news can produce outsized short squeezes. Q3 results involved a reduction in EPS guidance from $1.60-1.80 to $1.40-1.55 but also the best quarter ever in the company’s history in terms of cash flow. TPC generated $200m in FCF in the quarter. The surprise sent the shares 18% higher on the day and nearly 100% up from Q2. We think the reaction gives us a very interesting entry point on the short and believe the market has not looked properly at the drivers of cash generation.

The CEO was quick to tout the strong cash flow generation as a victory for him on the collection front, vindicating his earlier claims. The CEO said “The Company generated a new quarterly record $222.9 million of operating cash for the third quarter…The strong operating cash flow was driven by collections associated with certain dispute resolutions, as well as from the Company’s continued focus on improved working capital management… Our strong cash flow was driven by significant collections associated with several settlements”. The perception that TPC got a handle on collections sent the shares higher. Furthermore, the CEO indicated that 9 individual disputes (claims) totalling $257m will be resolved (either settled or via legal means) in Q1-2020. This further positive development reinforced the perception that collections are a problem of the past and significant cash flow will be generated in the near term. Even bears Gradient effectively threw in the towel on the short and upgraded the stock following these comments: “we note that a portion of our concern is scheduled to be resolved via litigation early next year”.

In our view, the market was blinded by the strong cash generation in the quarter, ignoring the real sources of cash. Contrary to what management would like you to believe, settlements of claims had very little to do with cash generation this past quarter. A per table below, the claims balance continued to climb higher each quarter this year:

Rather, what is driving cash flow, is the increase in “Billings in excess of costs and estimated earnings”, otherwise known as Deferred Revenues

Understanding movements in Deferred Revenues

Billings in excess of costs and estimated earnings, or deferred revenues, is defined as the excess of contract billings to date over the amount of contract costs and profits (or contract revenue) recognized to date. In other words, it represents the cash amount clients advanced to TPC for large project that wasn’t spent yet. It’s a sort of advance payment, typically 5-10% of the contract value, that clients pay contractors to help them managing their working capital. In a normally managed business, one would expect this balance to remain stable over time, at least as a proportion of revenues. There is no logical reason to explain why clients would decide to prepay larger proportion of projects over time. However, this is exactly what happened to TPC clients in recent years:

Until 2017, deferred revenue balance was in the $300-400m range, or 6-8% of sales. Over the last 2-3 years, the balance jumped to over $800m, or 18% of sales, nearly 3x larger than the “normal” level. This was a huge source of cash flow for TPC in recent periods. Since Q1-17, TPC raised nearly $500m in cash from changes in deferred revenue: $126m in 2017, $116m in 2018 and $246m in the 9 months to September 2019. It’s fair to say that without this huge increase in deferred, TPC would probably be bust by now.

What is driving this huge increase in Deferred? This is probably the most interesting and misunderstood element of TPC business. We think that TPC is desperate for cash, they cannot collect receivables and the only way to fund themselves is to ask clients to do it for them. We think TPC is signing as many projects as they can because every time a new project is signed, the client will have to put down some 5-10% of the total project cost up front, thereby advancing working capital to an otherwise stressed balance sheet. We think TPC is effectively committing themselves to a huge amount of liabilities (remember – at least 80% of contracts are fixed price, if there is a cost overrun or the contract was bid overly aggressively, TPC is on the hook for the losses) in order to get some cash in the door. This is also very clear just by looking at the backlog. Backlog ballooned to $11bn from $6bn 3 years ago while sales are still lower than they were then.

TPC would justify this by saying that a large backlog means higher revenues in the future. While this may well still happen in the future, the huge increase in backlog didn’t bring any revenue growth over the last 4 years:

Let’s think about this for a second. Over the last 4 years, revenues went nowhere but backlog went up by over $4bn. We also know that over the same period, deferred revenues increased by c. $0.5bn. A typical contract has c. 10% down payment / advance payment. Signing incremental $4-5bn of backlog should lead to $400-500m of cash inflow from deferred. This is exactly what TPC did – they signed as many contracts as they could in order to get hold of the cash.

How will the story unfold?

We think that TPC is playing a dangerous game that could lead to its own demise. Signing contracts to get advance payments is great, but what TPC is doing is effectively signing liabilities. They will need to deliver and perform on these contracts, or otherwise be subject to financial penalties. We believe there are 2 fundamental issues TPC will need to deal with:

1. While we have no proof of this, we suspect these contracts carry a huge amount of risk to TPC and many of them are likely to be unprofitable. We base this hypothesis on the following 4 considerations:
   1. Past contracts were clearly unprofitable at a cash flow level, so there is no reason to suspect these are going to be any better
   2. The proportion of fixed price contracts increased to 80% from less than 50% in 2016, suggesting TPC is willing to take on a lot more risk
   3. TPC needs to sign as many contracts as they can for working capital reasons, as explained above. It is likely that in order to win these competitive bids, they had to bid rather aggressively
   4. TPC financial profile is questionable. Its bond yields 7%. All else being equal, clients would rather advance the cash to a financially sound competitor. TPC may have had to underbid in order to win a tender
2. Even assuming the contract signed are profitable, TPC will need to unwind the positive working capital effect from advances from customers. Unless backlog continues to increase and they continue to lock themselves in ever increasing larger future liabilities, the deferred revenue balance will inevitably shrink. These contracts will eventually need to be worked on. This is going to be a huge use of cash for TPC going forward

It’s very hard to predict how and when TPC will fold but we think the issues described above are huge and there are no easy fixes. We believe the only way to restore a healthy balance sheet would be to raise a substantial amount of fresh equity in a highly dilutive capital increase.

Conclusion

We have no smoking gun. We can only highlight several yellow flags that taken together paint a picture of a company with a business vastly underperforming reported financials while carrying a substantially stressed balance sheet and no easy way out of this mess.

We could be wrong but the risk-reward appears very attractive. Even assuming TPC will be able to collect overdue receivables and continue to fund its operations, current valuation doesn’t leave much room to the upside. Over the last decade, TPC generated on average c. $220-260m in EBITDA. Even assuming these numbers are real (and we think they are not given aggressive accounting issues), the current share price implies a valuation multiple of 6-7x EV/EBITDA, which is at the high end of where peers are trading (GVA is on 6x, FLR on 5x, BBY LN on 6.5x). If, on the other hand, our bear case plays out, the equity could be worthless.

Catalyst

It’s very hard to predict how and when TPC will fold but we think the issues described above are huge and there are no easy fixes. We believe the only way to restore a healthy balance sheet would be to raise a substantial amount of fresh equity in a highly dilutive capital increase.